Deep Natural Language Processing for Search and Recommender SystemsHuiji Gao
Tutorial for KDD 2019:
Search and recommender systems process rich natural language text data such as user queries and documents. Achieving high-quality search and recommendation results requires processing and understanding such information effectively and efficiently, where natural language processing (NLP) technologies are widely deployed. In recent years, the rapid development of deep learning models has been proven successful for improving various NLP tasks, indicating their great potential of promoting search and recommender systems.
In this tutorial, we summarize the current effort of deep learning for NLP in search/recommender systems. We first give an overview of search/recommender systems with NLP, then introduce basic concept of deep learning for NLP, covering state-of-the-art technologies in both language understanding and language generation. After that, we share our hands-on experience with LinkedIn applications. In the end, we highlight several important future trends.
A Multi-Armed Bandit Framework For Recommendations at NetflixJaya Kawale
In this talk, we present a general multi-armed bandit framework for recommendations on the Netflix homepage. We present two example case studies using MABs at Netflix - a) Artwork Personalization to recommend personalized visuals for each of our members for the different titles and b) Billboard recommendation to recommend the right title to be watched on the Billboard.
Personalized Page Generation for Browsing RecommendationsJustin Basilico
Talk from First Workshop on Recommendation Systems for TV and Online Video at RecSys 2014 in Foster City, CA on 2014-10-10 about how we personalize the layout of the Netflix homepage to make it easier for people to browse the recommendations to quickly find something to watch and enjoy.
Deep Learning in Recommender Systems - RecSys Summer School 2017Balázs Hidasi
This is the presentation accompanying my tutorial about deep learning methods in the recommender systems domain. The tutorial consists of a brief general overview of deep learning and the introduction of the four most prominent research direction of DL in recsys as of 2017. Presented during RecSys Summer School 2017 in Bolzano, Italy.
Shallow and Deep Latent Models for Recommender SystemAnoop Deoras
In this presentation, we survey latent models, starting with shallow and progressing towards deep, as applied to personalization and recommendations. After providing an overview of the Netflix recommender system, we discuss research at the intersection of deep learning, natural language processing and recommender systems and how they relate to traditional collaborative filtering techniques. We will present case studies in the space of deep latent variable models applied to recommender systems.
Deep learning based recommender systems (lab seminar paper review)hyunsung lee
This document summarizes different types of deep learning techniques used for recommender systems, including MLP, autoencoders, CNNs, RNNs, and DSSM. It discusses how each technique can help address challenges like exploiting complex side information, alleviating cold-start problems, and modeling temporal dynamics. Some examples of papers applying these techniques are provided. The document concludes by discussing potential future research directions, such as better modeling users/items, temporal recommendations, multi-task learning across domains, and new evaluation metrics.
Personalizing "The Netflix Experience" with Deep LearningAnoop Deoras
These are the slides from my talk presented at AI Next Con conference in Seattle in Jan 2019. Here I talk in a bit more detail about the intuition behind collaborative filtering and go a bit deeper into the details of non linear deep learned models.
The Netflix experience is driven by a number of Machine Learning algorithms: personalized ranking, page generation, search, similarity, ratings, etc. On the 6th of January, we simultaneously launched Netflix in 130 new countries around the world, which brings the total to over 190 countries. Preparing for such a rapid expansion while ensuring each algorithm was ready to work seamlessly created new challenges for our recommendation and search teams. In this post, we highlight the four most interesting challenges we’ve encountered in making our algorithms operate globally and, most importantly, how this improved our ability to connect members worldwide with stories they'll love.
In this lecture, I will first cover the recent advances in neural recommender systems such as autoencoder-based and MLP-based recommender systems. Then, I will introduce the recent achievement for automatic playlist continuation in music recommendation.
This document discusses an activity ranking system for LinkedIn's feed. It describes LinkedIn's large, heterogeneous set of over 40 activity types that must be ranked. It considers straw man approaches like reverse chronological ordering or ranking by social popularity, but notes these do not ensure relevance. The system models each activity with features of the actor, viewer, and their relationship. A machine learning model is trained on a large scale to predict click-through rates and rank activities. The system considers features like freshness, personalization, and past impressions. Bucket testing showed the approach improved over baselines by considering factors like freshness, personalization, and different platforms.
Past, Present & Future of Recommender Systems: An Industry PerspectiveJustin Basilico
Slides from our talk at the RecSys 2016 conference in Boston, MA 2016-09-18 on our perspective for what are important areas for future work in recommender systems.
ML Infra for Netflix Recommendations - AI NEXTCon talkFaisal Siddiqi
Faisal Siddiqi presented on machine learning infrastructure for recommendations. He outlined Boson and AlgoCommons, two major ML infra components. Boson focuses on offline training for both ad-hoc exploration and production. It provides utilities for data preparation, feature engineering, training, metrics, and visualization. AlgoCommons provides common abstractions and building blocks for ML like data access, feature encoders, predictors, and metrics. It aims for composability, portability, and avoiding training-serving skew.
This document proposes a calibrated recommendations approach that aims to provide recommendations that reflect all of a user's interests in correct proportions. Standard recommender systems trained for accuracy can lead to unbalanced recommendations that amplify a user's main interests and crowd out lesser interests. The calibrated recommendations approach uses a post-processing re-ranking step to optimize a submodular calibration metric, balancing accuracy and fairness by recommending items from all a user's interests in their correct proportions. Experiments on MovieLens data show that calibration can be improved significantly without degrading accuracy much.
Overview of the Recommender system or recommendation system. RFM Concepts in brief. Collaborative Filtering in Item and User based. Content-based Recommendation also described.Product Association Recommender System. Stereotype Recommendation described with advantage and limitations.Customer Lifetime. Recommender System Analysis and Solving Cycle.
Data council SF 2020 Building a Personalized Messaging System at NetflixGrace T. Huang
This document discusses building a personalized messaging system at Netflix to recommend content to users. It covers four key considerations:
1) Personalizing messaging decisions using classification techniques like logistic regression on outcome features.
2) Removing bias from the system using techniques like Thompson sampling, exploration-exploitation, and propensity correction.
3) Maximizing causal impact by explicitly modeling past actions and comparing member satisfaction with and without messages.
4) Balancing reward against cost by imposing a volume constraint like an incrementality threshold and using reinforcement learning approaches.
Counterfactual Learning for RecommendationOlivier Jeunen
Slides for our presentation at the REVEAL workshop for RecSys '19 in Copenhagen and a Data Science Leuven Meetup, titled "Counterfactual Learning for Recommendation".
Tutorial: Context In Recommender SystemsYONG ZHENG
This document provides an overview of a tutorial on context-aware recommender systems. The tutorial will cover traditional recommendation techniques, context-aware recommendation which incorporates additional contextual information such as time and location, and context suggestion. It includes an agenda with topics, background information on recommender systems and evaluation metrics, and descriptions of techniques for context-aware recommendation including context filtering and modeling.
Recommender system algorithm and architectureLiang Xiang
1) The document discusses recommender system algorithms and architecture. It covers common recommendation techniques like collaborative filtering, content-based filtering, and graph-based recommendations.
2) It also discusses challenges like cold starts for new users and items. For new users, it recommends using demographic data or initial feedback to understand interests. For new items, it suggests using content information or initial user feedback.
3) The document proposes a feature-based recommendation framework that connects users, items, and latent features to address challenges like heterogeneous data and cold starts. This framework provides explanations but does not support user-based methods.
Learning to Rank for Recommender Systems - ACM RecSys 2013 tutorialAlexandros Karatzoglou
The slides from the Learning to Rank for Recommender Systems tutorial given at ACM RecSys 2013 in Hong Kong by Alexandros Karatzoglou, Linas Baltrunas and Yue Shi.
Recurrent Neural Networks have shown to be very powerful models as they can propagate context over several time steps. Due to this they can be applied effectively for addressing several problems in Natural Language Processing, such as Language Modelling, Tagging problems, Speech Recognition etc. In this presentation we introduce the basic RNN model and discuss the vanishing gradient problem. We describe LSTM (Long Short Term Memory) and Gated Recurrent Units (GRU). We also discuss Bidirectional RNN with an example. RNN architectures can be considered as deep learning systems where the number of time steps can be considered as the depth of the network. It is also possible to build the RNN with multiple hidden layers, each having recurrent connections from the previous time steps that represent the abstraction both in time and space.
This document provides a tutorial on recurrent neural networks (RNNs) and long short-term memory (LSTM) networks. It begins with introductions to the speaker and an overview of the content. It then explains RNNs and how they work sequentially through hidden layers. Issues like vanishing gradients are discussed. LSTMs are introduced as an advanced RNN that can retain information over longer periods of time using gates. Pre-trained word embeddings like Word2Vec, GloVe, and FastText are briefly explained. Finally, homework is assigned to build a sentiment analysis model using an LSTM and pre-trained word embeddings on a Chinese text dataset.
Artificial Intelligence, Machine Learning and Deep LearningSujit Pal
Slides for talk Abhishek Sharma and I gave at the Gennovation tech talks (https://gennovationtalks.com/) at Genesis. The talk was part of outreach for the Deep Learning Enthusiasts meetup group at San Francisco. My part of the talk is covered from slides 19-34.
Deep dive into the world of word vectors. We will cover - Bigram model, Skip-gram, CBOW, GLO. Starting from simplest models, we will journey through key results and ideas in this area.
https://www.meetup.com/Deep-Learning-Bangalore/events/239996690/
This document provides an overview and introduction to representation learning of text, specifically word vectors. It discusses older techniques like bag-of-words and n-grams, and then introduces modern distributed representations like word2vec's CBOW and Skip-Gram models as well as the GloVe model. The document covers how these models work, are evaluated, and techniques to speed them up like hierarchical softmax and negative sampling.
Hardware Acceleration for Machine LearningCastLabKAIST
This document provides an overview of a lecture on hardware acceleration for machine learning. The lecture will cover deep neural network models like convolutional neural networks and recurrent neural networks. It will also discuss various hardware accelerators developed for machine learning, including those designed for mobile/edge and cloud computing environments. The instructor's background and the agenda topics are also outlined.
This document provides an introduction to deep learning. It discusses key concepts such as neural network layers that process input tensors, common layer types like convolutional and recurrent layers, and how networks are trained using stochastic gradient descent. Examples of deep learning applications that have achieved near-human level performance are also presented, such as image classification and speech recognition. The document then focuses on convolutional neural networks, covering concepts like convolution operations, spatial hierarchies, and max pooling. It concludes with a demonstration of digit and X-ray image classification using Keras and techniques for dealing with overfitting like dropout and data augmentation.
Deep Learning Introduction - WeCloudDataWeCloudData
This document provides an overview of machine learning and deep learning concepts including:
- Machine learning basics such as supervised vs. unsupervised learning and performance measures.
- A brief history of deep learning and basics such as neural networks.
- Linear algebra concepts from vectors to tensors that are important for machine learning.
- Specific machine learning algorithms including linear regression, logistic regression, and TensorFlow basics for defining and executing computation graphs.
Deep Learning & NLP: Graphs to the Rescue!Roelof Pieters
This document provides an overview of deep learning and natural language processing techniques. It begins with a history of machine learning and how deep learning advanced beyond early neural networks using methods like backpropagation. Deep learning methods like convolutional neural networks and word embeddings are discussed in the context of natural language processing tasks. Finally, the document proposes some graph-based approaches to combining deep learning with NLP, such as encoding language structures in graphs or using finite state graphs trained with genetic algorithms.
Machine Learning, Deep Learning and Data Analysis IntroductionTe-Yen Liu
The document provides an introduction and overview of machine learning, deep learning, and data analysis. It discusses key concepts like supervised and unsupervised learning. It also summarizes the speaker's experience taking online courses and studying resources to learn machine learning techniques. Examples of commonly used machine learning algorithms and neural network architectures are briefly outlined.
[Paper Reading] Attention is All You NeedDaiki Tanaka
The document summarizes the "Attention Is All You Need" paper, which introduced the Transformer model for natural language processing. The Transformer uses attention mechanisms rather than recurrent or convolutional layers, allowing for more parallelization. It achieved state-of-the-art results in machine translation tasks using techniques like multi-head attention, positional encoding, and beam search decoding. The paper demonstrated the Transformer's ability to draw global dependencies between input and output with constant computational complexity.
Artificial Intelligence and Deep Learning in Azure, CNTK and TensorflowJen Stirrup
Artificial Intelligence and Deep Learning in Azure, using Open Source technologies CNTK and Tensorflow. The tutorial can be found on GitHub here: https://github.com/Microsoft/CNTK/tree/master/Tutorials
and the CNTK video can be found here: https://youtu.be/qgwaP43ZIwA
A comprehensive tutorial on Convolutional Neural Networks (CNN) which talks about the motivation behind CNNs and Deep Learning in general, followed by a description of the various components involved in a typical CNN layer. It explains the theory involved with the different variants used in practice and also, gives a big picture of the whole network by putting everything together.
Next, there's a discussion of the various state-of-the-art frameworks being used to implement CNNs to tackle real-world classification and regression problems.
Finally, the implementation of the CNNs is demonstrated by implementing the paper 'Age ang Gender Classification Using Convolutional Neural Networks' by Hassner (2015).
The document discusses deep learning techniques. It introduces deep learning and its impact on fields like speech recognition and computer vision. Deep learning uses large datasets and high-capacity models with many parameters to exploit information in data more effectively than traditional machine learning. The document also discusses key developments in deep learning like increased data, deeper network architectures, and accelerated training with GPUs. It provides examples of common neural network architectures, losses, and activation functions used in deep learning.
Learn to Build an App to Find Similar Images using Deep Learning- Piotr TeterwakPyData
This document discusses using deep learning and deep features to build an app that finds similar images. It begins with an overview of deep learning and how neural networks can learn complex patterns in data. The document then discusses how pre-trained neural networks can be used as feature extractors for other domains through transfer learning. This reduces data and tuning requirements compared to training new deep learning models. The rest of the document focuses on building an image similarity service using these techniques, including training a model with GraphLab Create and deploying it as a web service with Dato Predictive Services.
Hanjun Dai, PhD Student, School of Computational Science and Engineering, Geo...MLconf
Graph Representation Learning with Deep Embedding Approach:
Graphs are commonly used data structure for representing the real-world relationships, e.g., molecular structure, knowledge graphs, social and communication networks. The effective encoding of graphical information is essential to the success of such applications. In this talk I’ll first describe a general deep learning framework, namely structure2vec, for end to end graph feature representation learning. Then I’ll present the direct application of this model on graph problems on different scales, including community detection and molecule graph classification/regression. We then extend the embedding idea to temporal evolving user-product interaction graph for recommendation. Finally I’ll present our latest work on leveraging the reinforcement learning technique for graph combinatorial optimization, including vertex cover problem for social influence maximization and traveling salesman problem for scheduling management.
Machine learning techniques can be applied in formal verification in several ways:
1) To enhance current formal verification tools by automating tasks like debugging, specification mining, and theorem proving.
2) To enable the development of new formal verification tools by applying machine learning to problems like SAT solving, model checking, and property checking.
3) Specific applications include using machine learning for debugging and root cause identification, learning specifications from runtime traces, aiding theorem proving by selecting heuristics, and tuning SAT solver parameters and selection.
Similar to Deep Learning for Personalized Search and Recommender Systems (20)
Reciprocating Air Compressor and its TypesAtif Razi
Air Compressors
Classification of Air Compressors
Reciprocating Air Compressor
Main Parts of Reciprocating Air Compressor
Working of Reciprocating Air Compressor
Types of Reciprocating Air Compressor
Applications of Reciprocating Air Compressor
Advantages & Disadvantages of Reciprocating Air Compressor
buy a fake University of London diploma supplementGlethDanold
Website: https://www.fakediplomamaker.shop/
Email: diplomaorder2003@gmail.com
Telegram: @fakeidiploma
skype: diplomaorder2003@gmail.com
wechat: jasonwilliam2003
buy bachelor degree from https://www.fakediplomamaker.shop/ to be competitive. Even if you are not already working and you havve just started to explore employment opportunities buy UK degree, buy masters degree from USA, buy bachelor degree from Australia, fake Canadian diploma where to buy diploma in Canada, It's still a great idea to purchase your degree and get a head start in your career. While many of the people your age will enlist in traditional programs and spend years learning you could accumulate valuable working experience. By the time they graduate you will have already solidified a respectable resume boasting both qualification and experience.
Structural Dynamics and Earthquake Engineeringtushardatta
Slides are prepared with a lot of text material to help young teachers to teach the course for the first time. This also includes solved problems. This can be used to teach a first course on structural dynamics and earthquake engineering. The lecture notes based on which slides are prepared are available in SCRIBD.
Computer Vision and GenAI for Geoscientists.pptxYohanes Nuwara
Presentation in a webinar hosted by Petroleum Engineers Association (PEA) in 28 July 2023. The topic of the webinar is computer vision for petroleum geoscience.
Artificial Intelligence Imaging - medical imagingNeeluPari
10 stages of Artificial Intelligence,
Artificial intelligence (AI) has made significant advancements in the field of medical imaging, offering valuable tools and capabilities to improve diagnostics, treatment planning, and patient care. Here are several ways AI is used in medical imaging
Deep Learning for Personalized Search and Recommender Systems
1. Deep Learning for Personalized
Search and Recommender
Systems
Ganesh Venkataraman
Airbnb
Nadia Fawaz, Saurabh Kataria, Benjamin Le, Liang Zhang
LinkedIn
1
2. Tutorial Outline
• Part I (45min) Deep Learning Key concepts
• Part II (45min) Deep learning for Search and Recommendations at Scale
• Coffee break (30 min)
• Deep Learning Case Studies
• Part III (45min) Jobs You May Be Interested In (JYMBII) at LinkedIn
• Part IV (45min) Job Search at LinkedIn
Q&A at the end of each part
2
3. Motivation – Why Recommender Systems?
• Recommendation systems are everywhere. Some examples of impact:
• “Netflix values recommendations at half a billion dollars to the company”
[netflix recsys]
• “LinkedIn job matching algorithms to improves performance by 50%” [San Jose
Mercury News]
• “Instagram switches to using algorithmic feed” [Instagram blog]
3
4. Motivation – Why Search?
4
PERSONALIZED SEARCH
4
Query = “things to do in halifax”
Search view – this is a classic IR problem
Recommendations view – For this query,
what are the recommended results?
5. Why Deep Learning? Why now?
• Many of the fundamental algorithmic techniques have existed since
the 80s or before
2.5 Exobytes of data produced per
day
Or 530,000,000 songs
150,000,000 iPhones 5
6. Why Deep Learning?
Image classification
eCommerce fraud
Search
Recommendations
NLP
Deep learning is eating the world
6
7. Why Deep Learning and Recommender
Systems?
• Features
• Semantic understanding of words/sentences possible with embeddings
• Better classification of images (identifying cats in YouTube videos)
• Modeling
• Can we cast matching problems into a deep (and possibly) wide net and learn
family of functions?
7
8. Part I – Representation Learning and Deep
Learning: Key Concepts
8
9. Deep Learning and AI
http://www.deeplearningbook.org/contents/intro.html 9
10. Part I Outline
• Shallow Models for Embedding Learning
• Word2Vec
• Deep Architectures
• FF, CNN, RNN
• Training Deep Neural Networks
• SGD, Backpropagation, Learning Rate Schedule, Regularization, Pre-Training
10
12. Representation learning for automated feature generation
• Natural Language Processing
• Word embedding: word2vec, GloVe
• Sequence modeling using RNN’s and LSTM’s
• Graph Inputs
• Deep Walk
• Multiple Hierarchy of features for varying granularities for semantic meaning
with deep networks
12
13. Example Application of Representation
Learning - Understanding Text
• One of the keys to any content based recommender system
is understanding text
• What does “understanding” mean?
• How similar/dissimilar are any two words?
• What does the word represent? (Named Entity
Recognition)
• “Abraham Lincoln, the 16th President ...”
• “My cousin drives a Lincoln”
13
14. How to represent a word?
• Vocabulary – run, jog, math
• Simple representation:
• [1, 0, 0], [0, 1, 0], [0, 0, 1]
• No representation of meaning
• Cooccurrence in a word/document matrix
14
15. How to represent a word?
• Trouble with cooccurrence matrix
• Large dimension, lots of memory
• Dimensionality reduction using SVD
• High computational time nxm matrix => O(mn^2)
• Adding new word => redo everything
15
16. Word embeddings taking context
• Key Conjecture
• Context matters.
• Words that convey a certain context occur together
• “Abraham Lincoln was the 16th President of the United States”
• Bigram model
• P (“Lincoln”|”Abraham”)
• Skip Gram Model
• Consider all words within context and ignore position
• P(Context|Word)
16
18. Word2Vec: Skip Gram Model
• Basic notations:
• w represents a word, C(w) represents all the context around a word
• 𝜃 represents the parameter space
• D represent all the (w, c) pairs
• 𝑝 𝑐 𝑤; 𝜃 represents the probability of context c given word w
parametrized by 𝜃
• The probability of all the context appearing given a word is given by:
• 𝑐∈𝐶(𝑤) 𝑝(𝑐|𝑤; 𝜃)
• The loss function then becomes:
• 𝑎𝑟𝑔𝑚𝑎𝑥 𝜃 𝑤,𝑐 ∈𝐷 𝑝(𝑐|𝑤; 𝜃)
18
19. Word2vec details
• Let 𝑣 𝑤 and 𝑣𝑐 represent the current word and context. Note that
𝑣𝑐 and 𝑣 𝑤 are parameters we want to learn
• p c w; 𝜃 =
𝑒 𝑣 𝑐∗𝑣 𝑤
𝑑∈𝐶 𝑒 𝑣 𝑑∗𝑣 𝑤
• C represents set of all available contexts
19
20. Negative Sampling – basic intuition
p c w; 𝜃 =
𝑒 𝑣 𝑐∗𝑣 𝑤
𝑑∈𝐶 𝑒 𝑣 𝑑∗𝑣 𝑤
• Sample from unigram distribution instead of taking all contexts into
account
• Word2vec itself is a shallow model and can be used to initialize a
deep model
20
22. Neuron: Computational Unit
• Input vector: x = [x1, x2 ,… ,xn]
• Neuron
• Weight vector: W
• Bias: b
• Activation function: f
• Output
a = f(WT x + b)
x1
x2
x3
x4
W
b
f
a = f(WTx + b)
Input x Neuron Output a 22
24. Layer
• Layer l: nl neurons
• weight matrix: W = [W1,…, Wnl]
• bias vector: b = [b1,…, bnl]
• activation function: f
• output vector
• a = f(WT x + b)
x1
x2
x3
x4
W1
b1
f
a1 = f(W1
T x + b1)
W2
b2
f
a2= f(W2
T x + b2)
Input x Layer Output a
W3
b3
f
a3= f(W3
T x + b3)
24
25. Layer: Matrix Notation
• Layer l: nl neurons
• weight matrix: W
• bias vector: b
• activation function: f
• output vector
• a = f(WT x + b)
• more compact notation
• fast-linear algebra routines for
quick computations in network
x1
x2
x3
x4
Input x Layer Output a
a = f(WT a + b)
W , b , f
25
27. Why CNN: Convolutional Neural Networks?
• Large size grid structured data
• 1D: time series
• 2D: image
• Convolution to extract features from image (e.g. edges, texture)
• Local connectivity
• Parameter sharing
• Equivariance to translation: small translations in input do not affect output
33. CNN example for image recognition: ImageNet [Krizhevsky et al., 2012]
Pictures courtesy of [Krizhevsky et al., 2012], http://www.cs.toronto.edu/~fritz/absps/imagenet.pdf
33
1st GPU
2nd GPU
filters learned by first CNN layer
34. Why RNN: Recurrent Neural Network?
• Sequential data processing
• ex: predict next word in sentence: “I was born in France. I can speak…”
• RNN
• Persist information through feedback loop
• loop passes information from one step to the next
• Parameter sharing across time indexes
• output unit depends on previous output units through same
update rule.
xt
ht
ht-1
35. Unfolded RNN
• Copies of NN passing feedback to one another
http://colah.github.io/posts/2015-08-Understanding-LSTMs/
35
36. LSTM: Long Short Term Memory [Hochreiter et al., 1997]
• Avoid vanishing or exploding gradient
• Cell state updates regulated by gates
• Forget: how much info from cell state to let
through
• Input: which cell state components to update
• Tanh: values to add to cell state
• Output: select component values to output
picture courtesy of http://colah.github.io/posts/2015-08-Understanding-LSTMs/
Cell state
• Long term dependencies
• large gap between relevant information and
where it is needed
• Cell state: long-term memory
• Can remember relevant information over long
period of time
36
37. Examples of RNN application
• Speech recognition [Graves et al., 2013]
• Language modeling [Mikolov, 2012]
• Machine translation [Kalchbrenner et al., 2013][Sustkever et al., 2014]
• Image captioning [Vinyals et al., 2014]
37
39. Cost Function
• m training samples (feature vector, label)
(𝑥 1 , 𝑦 1 ), … , (𝑥 𝑚 , 𝑦 𝑚 )
• Per sample cost: error between label and output from prediction layer
𝐽 𝑊, 𝑏; 𝑥 𝑖 , 𝑦 𝑖 = 𝑎(𝐿) 𝑥 𝑖 − 𝑦(𝑖) 2
• Minimize cost function over parameters: weights W and biases b
𝐽 𝑊, 𝑏 =
1
𝑚
𝑖=1
𝑚
𝐽(𝑊, 𝑏; 𝑥 𝑖
, 𝑦(𝑖)
) +
𝜆
2
𝑙=1
𝐿
𝑊(𝑙)
𝐹
2
Average error Regularization 39
40. Gradient Descent
• Random parameter initialization: symmetry breaking
• Gradient descent step: update for every parameter Wij
(l) and bi
(l)
𝜃 = 𝜃 − 𝛼𝛻θ 𝔼[𝐽(𝜃)]
• Gradient computed by Backpropagation
• High cost of backpropagation over full training set
40
41. Stochastic Gradient Descent (SGD)
• SGD: follow negative gradient after
• single sample
𝜃 = 𝜃 − 𝛼𝛻𝜃J(θ; 𝑥 𝑖
, 𝑦(𝑖)
)
• a few samples: mini-batch (256)
• Epoch: full pass through training set
• Randomly shuffle data prior to each training epoch
41
42. Backpropagation [Rumelhart et al., 1986]
Goal: Compute gradient numerically
Recursively apply chain rule for derivative of composition of functions
Let 𝑦 = 𝑔 𝑥 and 𝑧 = 𝑓 𝑦 = 𝑓(𝑔(𝑥))
then
𝜕𝑧
𝜕𝑥
=
𝜕𝑧
𝜕𝑦
𝜕𝑦
𝜕𝑥
= 𝑓′
𝑔 𝑥 𝑔′(𝑥)
Backpropagation steps
1. Feedforward pass: compute all activations
2. Output error: measures node contribution to output error
3. Backpropagate error through all layers
4. Compute partial derivatives
42
43. Training optimization
• Learning Rate Schedule
• Changing learning rate as learning progresses
• Pre-training
• Goal: training simple model on simple task before training desired model to perform desired task
• Greedy supervised pre-training: pre-train for task on subset of layers as initialization for final network
• Regularization to curb overfitting
• Goal: reduce generalization error
• Penalize parameter norm: L2, L1
• Augment dataset: train on more data
• Early stopping: return parameter set at point in time with lowest validation error
• Drop out [Srivatstava, 2013] : train ensemble of all subnetworks formed by removing non-output units
• Gradient clipping to avoid exploding gradient
• norm clipping
• element wise clipping
43
44. Part II – Deep Learning for Personalized
Recommender Systems at Scale
44
48. item j from a set of candidates
User i
with
<user features, query
(optional)>
(e.g., industry,
behavioral features,
Demographic features,……)
(i, j) : response yijvisits
Algorithm selects
(action or not, e.g. click, like, share, apply…)
Which item(s) should we recommend to the user?
• The item(s) with the best expected utility
• Utility examples:
• CTR, Revenue, Job Apply rates, Ads conversion rates, …
• Can be a combination of the above for trade-offs
Personalized Recommender Systems
48
50. User
Interaction
Logs
Offline Modeling
Workflow + User /
Item derived
features
User
User Feature
Store
Item Store +
Features
Recommendation
Ranking
Ranking
Model Store
Additional Re-
ranking Steps
1
2
4
5
Offline System Online System
3
An example of Recommender System
Architecture
Item
derived features
50
51. User
Interaction
Logs
Offline Modeling
Workflow + User /
Item derived
features
User
Search-based
Candidate
Selection &
Retrieval
Query
Construction
User Feature
Store
Search Index
of Items
Recommendation
Ranking
Ranking
Model Store
Additional Re-
ranking Steps
1
2
3
4 5
6
7
Offline System Online System
Item
derived features
An example of Personalized Search
System Architecture
51
52. Key Components – Offline Modeling
• Train the model offline (e.g. Hadoop)
• Push model to online ranking model store
• Pre-generate user / item derived features for online systems
to consume
• E.g. user / item embeddings from word2vec / DNNs based
on the raw features
52
53. Key Components – Candidate Selection
• Personalized Search (With user query):
• Form a query to the index based on user query annotation [Arya et al., 2016]
• Example: Panda Express Sunnyvale +restaurant:panda express
+location:sunnyvale
• Recommender system (Optional):
• Can help dramatically reduce the number of items to score in ranking steps
[Cheng, et al., 2016, Borisyuk et al. 2016]
• Form a query based on the user features
• Goal: Fetch only the items with at least some match with user feature
• Example: a user with title software engineer -> +title:software engineer for
jobs recommendation
53
54. Key Components - Ranking
• Recommendation Ranking
• The main ML model that ranks items retrieved by candidate selection based
on the expected utility
• Additional Re-ranking Steps
• Often for user experience optimization related to business rules, e.g.
• Diversification of the ranking results
• Recency boost
• Impression discounting
• …
54
55. Integration of Deep Learning Models
into Personalized Recommender
Systems at Scale
55
56. Literature: Deep Learning for Recommendation Systems
• RBM for Collaborative Filtering [Salakhutdinov et al., 2007]
• Deep Belief Networks [Hinton et al., 2006]
• Neural Autoregressive Distribution Estimator (NADE) [Zheng, 2016]
• Neural Collaborative Filtering [He, et al., 2017]
• Siamese networks for user item matching [Huang et al., 2013]
• Deep Belief Networks with Pre-training [Hinton et al., 2006]
• Collaborative Deep Learning [Wang et al., 2015]
56
57. User
Interaction
Logs
Offline Modeling
Workflow + User /
Item derived
features
User
Search-based
Candidate
Selection &
Retrieval
Query
Construction
User Feature
Store
Search Index
of Items
Recommendation
Ranking
Ranking
Model Store
Additional Re-
ranking Steps
1
2
3
4 5
6
7
Offline System Online System
Item
derived features
57
58. Offline Modeling + User / Item Embeddings
User Features Item Features
User Embedding
Vector
Item Embedding
Vector
Sim(U,I)
User Feature
Store
Item Store / Index
with Features
58
59. Query Formulation & Candidate Selection
• Issues of using raw text: Noisy or incorrect query tagging due to
• Failure to capture semantic meaning
• Ex. Query: Apple watch -> +food:apple +product:watch or +product:apple watch?
• Multilingual text
• Query: 熊猫快餐 -> +restaurant:panda express
• Cross-domain understanding
• People search vs job search
59
60. Query Formulation & Candidate Selection
• Represent Query as an
embedding
• Expand query to similar
queries in a semantic
space
• KNN search in dense
feature space with
Inverted Index [Cheng,
et al., 2016]
Q = “Apple Watch”
D = “iphone”
D = “Orange Swatch”
D = “ipad”
60
61. Recommendation Ranking Models
• Wide and Deep Models to capture all possible signals [Cheng, et
al., 2016]
https://arxiv.org/pdf/1606.07792.pdf
61
62. Challenges & Open Problems for Deep
Learning at Recommender Systems
• Distributed training on very large data
• Tensorflow on Spark (https://github.com/yahoo/TensorFlowOnSpark)
• CNTK (https://github.com/Microsoft/CNTK)
• MXNet (http://mxnet.io/)
• Caffe (http://caffe.berkeleyvision.org/)
• …
• Latency Issues from Online Scoring
• Pre-generation of user / item embeddings
• Multi-layer scoring (simple models => complex)
• Batch vs online training
62
63. Part III – Case Study: Jobs You May Be
Interested In (JYMBII)
63
64. Outline
• Introduction
• Generating Embeddings via Word2vec
• Generating Embeddings via Deep Networks
• Tree Feature Transforms in Deep + Wide Framework
64
66. Introduction: Problem Formulation
• Rank jobs by 𝑃 User 𝑢 applies to Job 𝑗 𝑢, 𝑗)
• Model response given:
66
Careers History, Skills, Education, Connections Job Title, Description, Location, Company
66
67. Introduction: JYMBII Modeling- Generalization
Recommend
• Model should learn general rules to predict which
jobs to recommend to a member.
• Learn generalizations based on similarity in title, skill,
location, etc between profile and job posting
67
68. Introduction: JYMBII Modeling - Memorization
Applies to
68
• Model should memorize exceptions to the rules
• Learn exceptions based on frequent co-
occurrence of features
69. Introduction: Baseline Features
• Dense BoW Similarity Features for Generalization
• i.e: Similarity in title text good predictor of response
• Sparse Two-Depth Cross Features for Memorization
• i.e: Memorize that computer science students will transition to entry engineering roles
Vector BoW Similarity Feature
Sim(User Title BoW,
Job Title BoW)
Sparse Cross Feature
AND(user = Comp Sci. Student,
job = Software Engineer)
Sparse Cross Feature
AND(user = In Silicon Valley,
job = In Austin, TX)
Sparse Cross Feature
AND(user = ML Engineer,
job = UX Designer)
69
70. Introduction: Issues
• BoW Features don’t capture semantic similarity between user/job
• Cosine Similarity between Application Developer and Software Engineer is 0
• Generating three-depth, four-depth cross features won’t scale
• i.e. Memorizing that Factory Workers from Detroit are applying to Fracking
jobs in Pennsylvania
• Hand-engineered features time consuming and will have low coverage
• Permutations of three-depth, four-depth cross features grows exponentially
70
71. Introduction: Deep + Wide for JYMBII
• BoW Features don’t capture semantic similarity between user/job
• Generate embeddings to capture Generalization through semantic similarity
• Deep + Wide model for JYMBII [Cheng et al., 2016]
Semantic Similarity Feature
Sim(User Embedding,
Job Embedding)
Global Model Cross Feature
AND(user = Comp Sci. Student,
job = Software Engineer)
User Model Cross Feature
AND(user = User 2,
job = Job Latent Feature 1 )
Job Model Cross Feature
AND(user = User Latent Feature,
job = Job 1)
71
Sparse Cross Feature
AND(user = Comp Sci. Student,
job = Software Engineer)
Sparse Cross Feature
AND(user = In Silicon Valley,
job = In Austin, TX)
Sparse Cross Feature
AND(user = ML Engineer,
job = UX Designer)
Vector BoW Similarity Feature
Sim(User Title BoW,
Job Title BoW)
72. Generating Embeddings via Word2vec:
Training Word Vectors
• Key Ideas
• Same users (context) apply to similar jobs (target)
• Similar users (target) will apply to the same jobs (context)
Application Developer => Software Engineer
• Train word vectors via word2vec skip-gram architecture
• Concatenate user’s current title and the applied job’s title as input
User Title Applied Job Title
72
73. Generating Embeddings via Word2vec:
Model Structure
Application, Developer Software, EngineerTokenized Titles
Word Embedding Lookup
Pre-trained Word
Vectors
Entity Embeddings
Via Average Pooling
Word Vectors
Response Prediction (Logistic Regression)
Cosine Similarity
User Job 73
74. Generating Embeddings via Word2vec:
Results and Next Steps
• Receiver Operating Characteristic – Area Under Curve for evaluation
• Response prediction is binary classification: Apply or don’t Apply
• Highly skewed data: Low CTR for Apply Action
• Good metric for ranking quality: Focus on discriminatory ability of model
• Marginal 0.87% ROC AUC Gain
• How to improve quality of embeddings?
• Optimize embeddings for prediction task with supervised training
• Leverage richer context about user and job
74
75. Generating Embeddings via Deep Networks:
Model Structure
User Job
Response Prediction (Logistic Regression)
Sparse Features (Title, Skill,
Company)
Embedding Layer
Hidden Layer
Entity Embedding
Hadamard Product (Elementwise Product)
75
76. Generating Embeddings via Deep Networks:
Hyper Parameters, Lots of Knobs!
• Optimizer Used
• SGD w/ Momentum and exponential decay vs. Adam [Kingma et al., 2015] (Adam)
• Learning Rate
• 10−5
to 10−3
(𝟏𝟎−𝟒
)
• Embedding Layer Size
• 50 to 200 (100)
• Dropout
• 0% to 50% dropout (0% dropout)
• Sharing Parameter Space for both user/job embeddings
• Assumes communitive property of recommendations (a + b = b + a) (No shared parameter space)
• Hidden Layer Sizes
• 0 to 2 Hidden Layers (200 -> 200 Hidden Layer Size)
• Activation Function
• ReLU vs. Tanh (ReLU)
76
77. Generating Embeddings via Deep Networks:
Training Challenges
• Millions of rows of training data impossible to store all in memory
• Stream data incrementally directly from files into a fixed size example pool
• Add shuffling by randomly sampling from example pool for training batches
• Extreme dimensionality of company sparse feature
• Reduce dimensionality of company feature from millions -> tens of thousands
• Perform feature selection by frequency in training set
• Hyper parameter tuning
• Distribute grid search through parallel modeling in single driver Spark jobs
77
78. Generating Embeddings via Deep Networks:
Results
Model ROC AUC
Baseline Model 0.753
Deep + Wide Model 0.790 (+4.91%***)
*** For reference, a previous major JYMBII
modeling improvement with a 20% lift in ROC
AUC resulted in a 30% lift in Job Applications
78
79. Response Prediction (Logistic Regression)
The Current Deep + Wide Model
Deep Embedding Features (Feed Forward NN)
• Generating three-depth, four-depth cross features won’t scale
• Smart feature selection required
Wide Sparse Cross Features (Two-Depth)
79
80. Tree Feature Transforms: Feature Selection via
Gradient Boosted Decision Trees
Each tree outputs a path from root to leaf encoding
a combination of feature crosses [He et al., 2014]
GDBT’s select the most useful combinations of
feature crosses for memorization
Member Seniority: Vice
President
Yes
No
Member Industry:
Banking
Yes
No
Member Location:
Silicon Valley
Member Skill:
Statistics
Yes No
80
Yes No
Job Seniority:
CXO
NoYes
Job Title: ML
Engineer
Yes No
81. Response Prediction (Logistic Regression)
Tree Feature Transforms: The Full Picture
How to train both the NN model and GBDT model
jointly with each other?
Deep Embedding Features (Feed Forward NN) Wide Sparse Cross Features (GBDT)
81
82. Tree Feature Transforms: Joint Training via
Block-wise Cyclic Coordinate Descent
• Treat NN model and GBDT model as separate block-wise coordinates
• Implemented by
1. Training the NN until convergence
2. Training GBDT w/ fixed NN embeddings
3. Training the regression layer weights w/ generated cross features from GBDT
4. Training the NN until convergence w/ fixed cross features
5. Cycle step 2-4 until global convergence criteria
82
83. Response Prediction (Logistic Regression)
Tree Feature Transforms: Train NN Until
Convergence
Initially no trees are in our forest
Deep Embedding Features (Feed Forward NN) Wide Sparse Cross Features (GDBT)
83
84. Response Prediction (Logistic Regression)
Tree Feature Transforms: Train GDBT w/ NN
Section as Initial Margin
Deep Embedding Features (Feed Forward NN) Wide Sparse Cross Features (GDBT)
84
85. Response Prediction (Logistic Regression)
Tree Feature Transforms: Train GDBT w/ NN
Section as Initial Margin
Deep Embedding Features (Feed Forward NN) Wide Sparse Cross Features (GDBT)
85
86. Response Prediction (Logistic Regression)
Tree Feature Transforms: Train Regression
Layer Weights
Deep Embedding Features (Feed Forward NN) Wide Sparse Cross Features (GDBT)
86
87. Response Prediction (Logistic Regression)
Tree Feature Transforms: Train NN w/ GDBT
Section as Initial Margin
Deep Embedding Features (Feed Forward NN) Wide Sparse Cross Features (GDBT)
87
88. Tree Feature Transforms: Block-wise
Coordinate Descent Results
Model ROC AUC
Baseline Model 0.753
Deep + Wide Model 0.790 (+4.91%)
Deep + Wide Model w/ GBDT Iteration 1 0.792 (+5.18%)
Deep + Wide Model w/ GBDT Iteration 2 0.794 (+5.44%)
Deep + Wide Model w/ GBDT Iteration 3 0.795 (+5.57%)
Deep + Wide Model w/ GBDT Iteration 4 0.796 (+5.71%)
88
89. JYMBII Deep + Wide: Future Direction
• Generating Embeddings w/ LSTM Networks
• Leverage sequential career history data
• Promising results in NEMO: Next Career Move Prediction with Contextual
Embedding [Li et al., 2017]
• Semi-Supervised Training
• Leverage pre-trained title, skill, and company embeddings on profile data
• Replace Hadamard Product for entity embedding similarity function
• Deep Crossing [Shan et al., 2016]
• Add even richer context
• i.e. Location, Education, and Network features
89
90. Part IV – Case Study: Deep Learning Networks
for Job Search
90
94. Introduction: Query Understanding -
Segmentation and Tagging
• First divide the search query into
segments
• Tag query segments based on
recognized entity tags
Oracle
Java
Application Developer
Oracle
Java Application Developer
Query Segmentations
COMPANY = Oracle
SKILL = Java
TITLE = Application Developer
COMPANY = Oracle
TITLE = Java Application
Developer
Query Tagging
94
95. Introduction: Query Understanding –
Expansion
• Task of adding additional
synonyms/related entities to the
query to improve recall
• Current Approach: Curated dictionary
for common synonyms and related
entities
COMPANY = Oracle OR NetSuite OR
Taleo OR Sun Microsystems OR …
SKILL = Java OR Java EE OR J2EE
OR JVM OR JRE OR JDK …
TITLE = Application Developer OR
Software Engineer OR
Software Developer OR
Programmer …
Green – Synonyms
Blue – Related Entities
95
96. Introduction: Query Understanding - Retrieval
and Ranking
COMPANY = Oracle OR NetSuite OR Taleo OR
Sun Microsystems OR …
SKILL = Java OR Java EE OR J2EE OR JVM
OR JRE OR JDK …
TITLE = Application Developer OR
Software Engineer OR
Software Developer OR
Programmer …
Title
Title
Skills
Company
96
97. Introduction: Issues – Retrieval and Ranking
• Term retrieval has limitations
• Cross language retrieval
• Softwareentwickler Software developer
• Word Inflections
• Engineering Management Engineering Manager
• Query expansion via curated dictionary of synonyms is not scalable
• Expensive to refresh and store synonyms for all possible entities
• Heavy reliance on query tagging is not robust enough
• Novel title, skill, and company entities will not be tagged correctly
• Errors upstream propagates to poor retrieval and ranking
97
98. Introduction: Solution – Deep Learning for
Query and Document Representations
• Query and document representations
• Map queries and document text to vectors in semantic space
• Robust to Handle Out of Vocabulary words
• Term retrieval has limitations
• Query expansion via curated dictionary of synonyms is not scalable
• Map synonyms, translations and inflections to similar vectors in semantic space
• Term retrieval on cluster id or KNN based retrieval
• Heavy reliance on query tagging is not robust enough
• Compliment structured query representations with semantic representations
98
99. Representations via Word2vec:
Leverage JYMBII Work
• Key Ideas
• Similar users (context) apply to the same job (target)
• The same user (target) will apply to similar jobs (context)
Application Developer => Software Engineer
• Train word vectors via word2vec skip-gram architecture
• Concatenate user’s current title and the applied job’s title as input
User Title Applied Job Title
99
100. Representations via Word2vec:
Word2vec in Ranking
Application, Developer Software, EngineerTokenized Text
Word Embedding Lookup
Pre-trained Word
Vectors
Entity Embeddings
Via Average Pooling
Word Vectors
Learning to Rank Model (NDCG Loss)
Cosine Similarity
JobQuery 100
101. Representations via Word2vec:
Ranking Model Results
Model Normalized Cumulative
Discounted Gain@5 (NDCG@5)
CTR@5(%)
Baseline Model 0.582 +0.0%
Baseline Model + Word2Vec Feature 0.595 (+2.2%) +1.6%
101
102. Representations via Word2vec:
Optimize Embeddings for Job Search Use Case
• Leverage apply and click feedback to guide learning of embeddings
• Fine tune embeddings for task using supervised feedback
• Handle out of vocabulary words and scale to query vocabulary size
• Compared to JYMBII, query vocabulary is much larger and less well-formed
• Misspellings
• Word Inflections
• Free text search
• Need to make representations more robust for these free text queries
102
103. Robust Representations via DSSM:
Deep Structured Semantic Model [Huang et al., 2013]
Query Applied Job (Positive)
Application Developer Software EngineerRaw Text
#Ap, App, ppl… #So, Sof, oft…Tri-letter Hashing #Ha, Hai, air…
Hairdresser
Randomly Sampled
Applied Job (Negative)
Hidden Layer 3
Hidden Layer 2
Hidden Layer 1
Cosine Similarity
Softmax w/ Cross Entropy Loss
103
104. Robust Representations via DSSM:
Tri-letter Hashing
• Tri-letter Hashing Example
• Engineer -> #en, eng, ngi, gin, ine, nee, eer, er#
• Benefits of Tri-letter Hashing
• More compact Bag of Tri-letters vs. Bag of Words representation
• 700K Word Vocabulary -> 75K Tri-letters
• Can generalize for out of vocabulary words
• Tri-letter hashing robust to minor misspellings and inflections of words
• Engneer -> #en, eng, ngn, gne, nee, eer, er#
104
105. Robust Representations via DSSM:
Training Details
105
• Parameter Sharing Helps
• Better and faster convergence
• Model size is reduced
• Regularization
• L2 performs better than dropout
• Toolkit Comparisons (CNTK vs TensorFlow)
• CNTK: Faster convergence and better model quality
• TensorFlow: Easy to implement and better community support.
Comparative model quality
Training performance with/o parameter sharing
106. Robust Representations via DSSM:
Lessons in Production Environment
106
+ 100%
+ 70%
+ 40%
• Bottlenecks in Production
Environment
• Latency due to extra computation
• Latency due to GC activity
• Fat Jars in JVM environment
• Practical Lessons
• Avoid JVM Heap while serving the
model
• Caching most accessed entities’
embedding
107. Robust Representations via DSSM:
DSSM Qualitative Results
Software Engineer Data Mining LinkedIn Softwareentwickler
Engineer Software Data Miner Google Software
Software Engineers Machine Learning
Engineer
Software Engineers Software Engineer
Software Engineering Microsoft Research Software Engineer Engineer Software
For qualitative results, only top head queries are taken to analyze similarity to each other
107
108. Robust Representations via DSSM:
DSSM Metric Results
Model Normalized Cumulative
Discounted Gain@5 (NDCG@5)
CTR@5 Lift (%)
Baseline Model 0.582 +0.0%
Baseline Model + Word2Vec Feature 0.595 (+2.2%) +1.6%
Baseline Model + DSSM Feature 0.602 (+3.4%) +3.2%
108
109. Robust Representations via DSSM:
DSSM Future Direction
• Leverage Current Query Understanding Into DSSM Model
• Query tag entity information for richer context embeddings
• Query segmentation structure can be considered into the network design
• Deep Crossing for Similarity Layer [Shan et al., 2016]
• Convolutional DSSM [Shen et al., 2014]
109
110. Conclusion
• Recommender Systems and personalized search are very similar
problems
• Deep Learning is here to stay and can have significant impact on both
• Understanding and constructing queries
• Ranking
• Deep learning and more traditional techniques are *not* mutually
exclusive (hint: Deep + Wide)
110
113. Difference between parameter sharing in 1-D
convolution and RNN?
• CNN Kernel: output unit depends on small number of neighboring input units
through same kernel
• RNN update rule: output unit depends on previous output units through same
update rule. Deeper computational graph.
115. Introduction: JYMBII Modeling -
Personalization
Applies to
Recommend jobs that are similar to jobs user has
previously applied to
115
116. Introduction: JYMBII Modeling - Collaboration
Applies to
Recommend jobs that similar users have previous
applied to
116
117. Introduction: Generalized Linear Mixed
Models
• Mixture of linear models into an additive model [Zhang et al., 2016]
• Fixed Effect – Population Average Model
• Random Effects – Subject Specific Models
Response Prediction (Logistic Regression)
User 1
Random Effect Model
User 2
Random Effect Model
Personalization
Job 2
Random Effect Model
Job 1
Random Effect Model
Collaboration
Global Fixed Effect Model
Content-Based Similarity
117
118. Introduction: Features
• Dense Vector BoW Similarity Features in global model for Generalization
• i.e: Similarity in title text good predictor of response
• Sparse Cross Features in global,user, and job model for Memorization
• i.e: Memorize that computer science students will transition to entry engineering roles
Vector BoW Similarity Feature
Sim(User Title BoW,
Job Title BoW)
Global Model Cross Feature
AND(user = Comp Sci. Student,
job = Software Engineer)
User Model Cross Feature
AND(user = User 2,
job = Software Engineer)
Job Model Cross Feature
AND(user = Comp Sci. Student,
job = Job 1)
118
119. Introduction: GLMM Formulation
• Generalized Linear
Mixed Models
• Per-User Random Effect
• Per-Job Random Effect
𝑃 𝐴𝑝𝑝𝑙𝑦 𝑚, 𝑗) = 𝜎(𝐵𝑓𝑖𝑥𝑒𝑑 𝑋cos 𝑚,𝑗 , 𝑋 𝑚, 𝑋𝑗, 𝑋 𝑚,𝑗 + 𝐵 𝑚 𝑋cos 𝑚,𝑗 , 𝑋𝑗 + 𝐵𝑗 𝑋cos 𝑚,𝑗 , 𝑋 𝑚 )
Notation Meaning
Xcos(m, j) Dense Cosine Similarity Features for sample pair m,j.
(i.e. - Cosine similarity between title BoW)
Xm Sparse Features of user m. (i.e. – User is a software
engineer)
Xj Sparse Features of job j. (i.e. – Job is from company
LinkedIn)
Xm,j Sparse Cross product feature transformation of Xm,
Xj. (i.e. – User is software engineer and job is from
company LinkedIn)
Bfixed Fixed effect model coefficients.
Bm User model coefficients for user m.
Bj Job model coefficients for job j.
Fixed Effect – User-Job Affinity Per-User Model Per-Job Model 119
120. Introduction: Issues
• BoW Features don’t capture semantic similarity between user/job
• Cosine Similarity between Application Developer and Software Engineer is 0
• Sparse features in user and job specific models make it difficult to fit
• Current linear model is unable to share learning across similar titles
• Large number of sparse features does not scale to infrastructure
• Billions of User and Job Model Cross features
120
121. Introduction: Proposed Solution
• Learn dense semantic embeddings of user and job entities
• Integrate embeddings into GLMM model as a set of latent features
Semantic Similarity Feature
Sim(User Embedding,
Job Embedding)
Global Model Cross Feature
AND(user = Comp Sci. Student,
job = Software Engineer)
User Model Cross Feature
AND(user = User 2,
job = Job Latent Feature 1 )
Job Model Cross Feature
AND(user = User Latent Feature,
job = Job 1)
121
122. Generating Embeddings Via Deep Networks:
Data and Features
• 4 weeks (Training) and 1 week (Test) of Click Log Data
• 15M Applies
• 2.5M Dismisses
• 30M Skips
• 15M Random Negatives
• 5M Users
• 1M Jobs
• Input Features
• Sparse Standardized Title - |Title Taxonomy| = 25K
• Sparse Standardized Skills - |Skill Taxonomy| = 35K
• Sparse Standardized Company - |Companies| = 1M+
122
123. Generating Embeddings via Deep Networks:
GLMM Integration Results
Improved model performance w/ 10x
reduction in number of model parameters
Model ROC AUC
Baseline GLMM Model w/ Sparse Features 0.800
GLMM Model w/ Dense Embedding Features 0.811 (+1.38%)
123
125. Robust Representations via DSSM:
DSSM Training Details
• Training Data: 6 months of apply data from job search
• Training Tuples: (query, applied job)
• Network Size: (3 Layers:- 300, 300, 300)
• Parameter Sharing: Shared Parameters among query and job
• Activation Function: Tanh
125
126. References
• [Rumelhart et al., 1986] Learning representations by back-propagating errors, Nature 1986
• [Hochreiter et al., 1997] Long short-term memory, Neural computation 1997
• [LeCun et al., 1998] Gradient-based learning applied to document recognition, Proceedings of
the IEEE 1998
• [Krizhevsky et al., 2012] Imagenet classification with deep convolutional neural networks, NIPS
2012
• [Graves et al., 2013] Speech recognition with deep recurrent neural networks, ICASSP 2013
• [Mikolov, 2012] Statistical language models based on neural networks, PhD Thesis, Brno
University of Technology, 2012
• [Kalchbrenner et al., 2013] Recurrent continuous translation models, EMNLP 2013
• [Srivatstava, 2013] Improving neural networks with dropout, PhD Thesis, University of Toronto,
2013
• [Sustkever et al., 2014] Sequence to sequence learningg with neural networks, NIPS 2014
• [Vinyals et al., 2014] Show and tell: a neural image caption generator, Arxiv 2014
• [Zaremba et al., 2015] Recurrent Neural Network Regularization, ICLR 2015
126
127. References (continued)
• [Arya et al., 2016] Personalized Federated Search at LinkedIn, CIKM 2016
• [Cheng et al., 2016] Wide & Deep Learning for Recommender Systems, DLRS 2016
• [He et al., 2014] Practical Lessons from Predicting Clicks on Ads at Facebook, ADKDD 2014
• [Kingma et al., 2015] Adam: A Method for Stochastic Optimization, ICLR 2015
• [Huang et al., 2013] Learning Deep Structured Semantic Models for Web Search using Clickthrough Data, CIKM 2013
• [Li et al., 2017] NEMO: Next Career Move Prediction with Contextual Embedding, WWW 2017
• [Shan et al., 2016] Deep Crossing: Web-scale modeling without manually crafted combinatorial features, KDD 2016
• [Zhang et al., 2016] GLMix: Generalized Linear Mixed Models For Large-Scale Response Prediction, KDD 2016
• [Salakhutdinov et al., 2007] Restricted Boltzmann Machines for Collaborative Filtering, ICML 2007
• [Zheng, 2016] http://tech.hulu.com/blog/2016/08/01/cfnade.html
• [Hinton et al., 2006] A fast learning algorithm for deep belief nets, Neural Computations 2006
• [Wang et al., 2015] Collaborative Deep Learning for Recommender Systems , KDD 2015
• [He et al., 2017] Neural Collaborative Filtering, WWW 2017
• [Borisyuk et al. 2016]. CaSMoS: A Framework for Learning Candidate Selection Models over Structured Queries and
Documents, KDD 2016
127
Welcome to our tutorial on Deep Learning for Personalized Search and Recommender Systems
I am Ganesh and I work for Airbnb. I will be presenting this with my former colleagues from LinkedIn – Nadia, Ben and Liang. Saurabh could not make it
Between the 5 of us we have worked on various aspects of search/recommendations/machine learning and deep learning.
We will be releasing these slides online. We want this to be an interactive tutorial, so please feel free to interrupt any time.
But, just to get a sense of the audience, show of hands to how many of you are from academia?
How many from industry?
How many have worked on some sort of production deep learning system before?
It is a fairly long tutorial and we want to make sure we have your attention
As we have noticed we have a fairly diverse audience and we want to ensure that we cater to all your interests which may be a challenge.
We start off with some foundational topic in deep learning. One may consider these like understanding the lego pieces
We then move on to explaining deep learning for search and recommendations at scale, followed by case study at LinkedIn
TODO – add logistics here
In Recys 2015, Netflix mentioned that recommender systems add about half a billion dollars to the company
It is almost a given these ,days that you open up a site or an app and it presents an experience that is tailor made for you, knowing your behavior and your preferences whether or not you made them explicit
There’s always been this friendly turf between search and recommendations.
Take a look at this query. A search person would treat this as classic IR problem, a recommendation person would treat this as a set of recommendations returned by the search engine for your query. Well, these two world’s aren’t that different and we believe they actually got married via personalized search.
Personalized search is where each result is tailored to your own preferences and not necessarily same for everyone. There are of course various levels of personalization
That brings us to the logical next question? Why deep learning
If you talk with a mathematician, they would point out that a lot of foundational techniques have been around since the 80’s or even before. And they would be right.
So, why now?
Two major things need to happen for deep learning to really work.
Computing cost
Availability of data – lot’s of it
TODO – add GPU
In several domains, deep learning has already made a dent. These include self driving cars and machine translation and a few others just to name some
deep learning is a particular case of representation learning.
It learns additional layers of more abstract features.
This slide should motivate why understanding text is critical
Understanding text is critical to any recommender system that works with underlying text data (example – news recommendations, jobs recommendations etc)
What are the components of “understanding”?
-- Similar or dissimilar words
-- concept of similarity can range from true synonyms to more ‘fuzzy’ types
-- Entity the word represents (Named Entity Recognition)
-- “Abraham Lincoln, the 16th President”, “My cousin drives a Lincoln”
Explain why the function isn’t scalable
Note on negative sampling
shallow models to learn embeddings, such as word2vec, are often used as initialization in deeper network architectures.
sigmoid used to convert real to (0,1), for example into a probability.
Used in output layer with log loss (cross entropy) in optimization, rather than squared loss, to avoid vanishing gradient issue when it saturates
discouraged use in hidden layer because of saturation
tanh resembles identity close to 0, can be trained easily as long as activation remains small. otherwise, saturation makes gradient-based learning difficult.
ReLU, also know as positive part
piecewise linear, easy to optimize
does not saturate on the positive side.
Initialize ReLU with small positive bias to make it likely for the ReLU to be active for most inputs in the training set
FF also known as Multi Layer Perceptron MLP
goal of the network: approximate a function y = F(x)
- non-linearities
output prediction layer:
linear unit for linear regression
sigmoid for logistic regression
softmax for multinoulli multi-class classification
CNN automates filters and feature extraction from images that used to be highly tailored and manual
Equivariance to translation: small translations in input do not affect output
Large image NxN pixels: N^2 input units
Hidden layer: K features
Number of parameters: ~KN^2
2 GPs to train large deep neural net (layers split between the 2 GPUs)
5 CNN layers with pooling + 3 fully connected layers
ReLU
DropOut against overfitting
Large dataset with 1000 image categories
image, label, classification. top 4: correctly classified, bottom 4: misclassified
63% accuracy
correct label was in top 5 predicted 85% of time
RNN update rule: output unit depends on previous output units through same update rule. Deeper computational graph.
1-D CNN Kernel: output unit depends on small number of neighboring input units through same kernel
chain structure with repeating module (parameter sharing across time)
cell state: conveyor belt that runs through the chain and stores long term memory.
cell state has a linear self-loop that allows product of derivatives close to one, weights of the self-loop are control by gates.
Problem with gradient descent for standard RNNs: error gradients vanish exponentially quickly with the size of the time lag between important events.
gradients propagated over many stages tend to either vanish (often) or explode
sigmoid gate lives in [0,1]: controls amount of information that it lets through from each component of cell state
tanh: create a vector of candidate values to update the cell state or from the cell state to update the output
speech recognition: acoustic modeling for mapping from acoustic signal to sequences of word (phonetic state)
machine translation: LSTM trained on concatenation of source sentences and their translation [Sustkever et al., 2014]
maximum likelihood: negative log likelihood = cross entropy between data distribution p(x,y) and model distribution p(y|x)
if model distribution is gaussian p(y|x) = N(y; a^L(x), I), then cross entropy boils down to squared loss.
alpha: learning rate
Dataset augmentation for image: translate, rotate, scale images to generate new samples.
Dropout:
- Dropout operator corrupts the information carried by the units, forcing them to perform their immediate computations more robustly
for LSTM, apply dropout only the non-recurrent connections, to maintain memorization ability see [Zaremba et al., 2015]
Gradient norm clipping maintains the direction of the gradient.
During time interval t
User visits to the webpage
Some item inventory at our disposal
Decision problem: for each visit, select an item to display
each display generates a response
Goal: choose items to maximize expected clicks
- Introduce My Self
- Passive Job Seekers: Allow them to discover what is avaliable in the marketplace.
- Not Alot of Data for Passives, show them jobs that make the most sense to them given their current career history and experience
- Active Job Seekers: Reinforce their job seeking experience. Show them similar jobs that they applied to that they may have missed. Make sure they don't miss opportunities
- Powers alot of modules including jobs home, feed, email, ads
- Goal: Get people hired - Confirmed Hires
- Time Lag on signal
- Proxy in Total Job Applies
- Metric: Total Job Applies
- Optimize probability of Apply. Not View. Showing users popular/attractive jobs not as important as showing them actual good matches
- User, Job, Activity
- Leverage our rich context on a user and job
- Generalize rules that similarity in title is a good thing
- Memorize exceptions
- I.E. for people in ML who have replaced our jobs w/ ML itself, we need to move to design
- Example of features in each of the model. Generalization, on most of the time to share learning across examples in a linear model
- Sparse Features for memorization
- Need to choose good features to memorize
- Random effect sparse features model personal affinity
- No semantic similarity do well for the most part
- Difficult fitting because no shared learning
- Don't scale to infrastructure. Too many parameter updates via SGD costly to run w/o too much ROI. Alot of sparse features need to be computed online which makes for expensive serving
- Dimensionality reduction
- Learn dense embeddings for job and user entities. Semantic embeddings
- Key things are to defined context and target. You shall know a word by the company it keeps
- Leverage word2vec existing libraries which take sentences as input
- Random Effects are not shown here
- Cross Features are not shown here
- Word vectors trained seperately
- ROC is good
- Accuracy won't be as good metric
- Marginal Gain from ROC AUC over BoW
- Leverage our large amount of labeled data
- Two Towers
- 3 Layers
- One tower for embedding member
- One for Job
- One similarity layer
- Non Linear Interactions. Important Skills for a Job
- Adam is adaptive learning rate. Momentum
- Momentum good for smoothing out noisy gradient
- Decay is good so that we don't overshoot the optium
- Dropout wasn't good for us. Model wasn't complex enough given our amount of data
- Sharing parameter space doesn't make sense for us
- Similar to Tensorflow batch queue implementation
- Millions of parameters * 100 in embedding size
- No distributed training so just train multiple models in parallel. Tensorflow supports distributed training but not on YARN
- Do well just knowing the member's profile
- Deep model does better due to all the non linear interactions introduced
- Gets close to the Baseline GLMM model, despite there being no random effects in the deep model to handle personalization and collaboration
- Can see how this translates
- Visual representation
- Model Driven way like for deep model?
- Random effects not shown
- Output of each tree is a path used as a categorical feature
- GDBT learns to select the most useful combinations of feature crosses for memorization
- Example of features
- Vice Presidents in Banking are not a normal VP. Seniority level is lower than other VP's in comparision
- Data Engineers who don't know programming but know statistics are still very good matches for Data engineering jobs
- Can't run SGD through graph
- SGD to train network
- Greedy split on learning decision rules
- Coordinant Descent is single coordinate optimzation (Iterative 1 dimensional optimization fixing the other coordinants)
- Train deep model first because we want to learn generalizations before exceptions. Can't have exceptions if we don't have rules first
- Deep Model Initialization. Train Deep Model first since need to learn the generalizations first before learning the exceptions
- Logit gets uses as an intial prediction for GDBT to start boosting from
- Tree learns to generate features to help classify cases where the NN is wrong
-
- Train Regression Layer weights on the features generated by tree layer transformation
- GDBT generates better features than sparse. Iterative training improves slightly
- AUC might not be best way to visualize how tree features help given that they won't move the needle as much since they handle exceptions
- Profile Data. Word2vec type of networks w/ career history
- Hadamard product is easy to deploy in production. Only need to do a vector operations rather than a full matrix multiply
- Location good to learn interactions between industry and location.
- Education good for students
- Search Jobs active job searchers
- Personalized Component to Job Search
- We will focus on the query to job matching though
- Extracting structured information from a user's raw query
- Query Understanding most important part
- Talk about indexer first, then talk about user query
- Reference LinkedIn Economic graph and taxonomy
- Add structure to query
-- Exact terms matching for retrieval will miss very relevant documents containing similar terms
- Expansion allows us to retrieve those very relevant documents
- Curate dictionary through domain knowledge or knowledge base
- Can also mine synonyms through query logs
- Give enough example for it
Google -> LinkedIn -> Apple
- Leverage query understanding for matching phrases rather than terms. Phrase match with distance
- Leverage query tags to match w/ appropriate fields for higher precision (Job Application -> Application Developer)
- BoW features (Title, query title sim)
- Might not be handled in synonyms
- Curating is not scalable given all the possible inflections and languages around. Large dictionary will be expensive to store all the mappings
- If query tagging fails, then our retrieval and ranking will not be as good
- Pinterest as a Skill, Hedge Fund as a company
-
- Good representations that map to dense semantic space
- Robust to OOV words
- Robust to issues w/ current segmentation/query tagging
- Map on job and query side for embeddings
- Compliment them. Query tagging gives us good information about the phrases and handles ambiguity for us but potentially, taxonomy not might be refreshed often enough to keep up w/ the professional world
- Embeddings learned from query log data should reflect new entities automatically as we see more new data
Same slide as before
- Show differences
NDCG Normalized Discounted Cumulative Gain
- Describe NDCG
- Relative Ranking
- Penalized lower rankings for relevant results
- Normalized across query
- CTR@5 goes up
Emphasize OOV words. Job search vocabulary alot more different from JYMBII vocab
- Skills, companies, titles, can search anything
- Talk about tri letter hashing later. BoTrigrams vs. BoW
- Extra Negatives. 10 Negatives
- 3 layers
- Softmax layer, learn positives and negatives at the same time
- Triletter hashing achieves goal of robustness at OOV
- Parameter Sharing for Full Siamese
- Reuse more
- Some Regularization used. Didn't really do too much though. Parameter sharing already adds some form of regularization
- Tensorflow for the flexibility
- Latency, model size, model delivery was issue
- Store model off heap to reduce GC activity
- Cache most access entities for speed
- Emphasize head queries
- Even better lift
- Also try to use for candidate selection
Recommendations and Search are more in common that what people may realize.
Online phase of any recommender or search consists of two very board stages and deep learning has impact on both