Export Citations
Save this search
Please login to be able to save your searches and receive alerts for new content matching your search criteria.
- research-articleAugust 2017
Sparse Compositional Local Metric Learning
KDD '17: Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data MiningAugust 2017, Pages 1097–1104https://doi.org/10.1145/3097983.3098153Mahalanobis distance metric learning becomes an especially challenging problem as the dimension of the feature space p is scaled upwards. The number of parameters to optimize grows with space complexity of order O (p 2), making storage infeasible, ...
- research-articleAugust 2017
Evaluating U.S. Electoral Representation with a Joint Statistical Model of Congressional Roll-Calls, Legislative Text, and Voter Registration Data
KDD '17: Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data MiningAugust 2017, Pages 1205–1214https://doi.org/10.1145/3097983.3098151Extensive information on 3 million randomly sampled United States citizens is used to construct a statistical model of constituent preferences for each U.S. congressional district. This model is linked to the legislative voting record of the legislator ...
- research-articleAugust 2017
Small Batch or Large Batch?: Gaussian Walk with Rebound Can Teach
KDD '17: Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data MiningAugust 2017, Pages 1275–1284https://doi.org/10.1145/3097983.3098147Efficiency of large-scale learning is a hot topic in both academic and industry. The stochastic gradient descent (SGD) algorithm, and its extension mini-batch SGD, allow the model to be updated without scanning the whole data set. However, the use of ...
- research-articleAugust 2017
Learning from Labeled and Unlabeled Vertices in Networks
KDD '17: Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data MiningAugust 2017, Pages 1265–1274https://doi.org/10.1145/3097983.3098142Networks such as social networks, citation networks, protein-protein interaction networks, etc., are prevalent in real world. However, only very few vertices have labels compared to large amounts of unlabeled vertices. For example, in social networks, ...
- research-articleAugust 2017
Google Vizier: A Service for Black-Box Optimization
KDD '17: Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data MiningAugust 2017, Pages 1487–1495https://doi.org/10.1145/3097983.3098043Any sufficiently complex system acts as a black box when it becomes easier to experiment with than to understand. Hence, black-box optimization has become increasingly important as systems have become more complex. In this paper we describe Google ...
- research-articleAugust 2017
Peeking at A/B Tests: Why it matters, and what to do about it
KDD '17: Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data MiningAugust 2017, Pages 1517–1525https://doi.org/10.1145/3097983.3097992This paper reports on novel statistical methodology, which has been deployed by the commercial A/B testing platform Optimizely to communicate experimental results to their customers. Our methodology addresses the issue that traditional p-values and ...
- research-articleAugust 2017
Retrospective Higher-Order Markov Processes for User Trails
KDD '17: Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data MiningAugust 2017, Pages 1185–1194https://doi.org/10.1145/3097983.3098127Users form information trails as they checkin with a geolocation, rate items, or consume media. A common problem is to predict what a user might do next for the purposes of guidance, recommendation, or prefetching. Markov chains models have been widely ...
- research-articleAugust 2017
Decomposed Normalized Maximum Likelihood Codelength Criterion for Selecting Hierarchical Latent Variable Models
KDD '17: Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data MiningAugust 2017, Pages 1165–1174https://doi.org/10.1145/3097983.3098110We propose a new model selection criterion based on the minimum description length principle in a name of the decomposed normalized maximum likelihood criterion. Our criterion can be applied to a large class of hierarchical latent variable models, such ...
- research-articleAugust 2017
Tripoles: A New Class of Relationships in Time Series Data
- Saurabh Agrawal,
- Gowtham Atluri,
- Anuj Karpatne,
- William Haltom,
- Stefan Liess,
- Snigdhansu Chatterjee,
- Vipin Kumar
KDD '17: Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data MiningAugust 2017, Pages 697–706https://doi.org/10.1145/3097983.3098099Mining relationships in time series data is of immense interest to several disciplines such as neuroscience, climate science, and transportation. Traditional approaches for mining relationships focus on discovering pair-wise relationships in the data. ...
- research-articleAugust 2017
Collaborative Variational Autoencoder for Recommender Systems
KDD '17: Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data MiningAugust 2017, Pages 305–314https://doi.org/10.1145/3097983.3098077Modern recommender systems usually employ collaborative filtering with rating information to recommend items to users due to its successful performance. However, because of the drawbacks of collaborative-based methods such as sparsity, cold start, etc., ...
- research-articleAugust 2017
Local Higher-Order Graph Clustering
KDD '17: Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data MiningAugust 2017, Pages 555–564https://doi.org/10.1145/3097983.3098069Local graph clustering methods aim to find a cluster of nodes by exploring a small region of the graph. These methods are attractive because they enable targeted clustering around a given seed node and are faster than traditional global graph clustering ...
- research-articleAugust 2017
Anomaly Detection with Robust Deep Autoencoders
KDD '17: Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data MiningAugust 2017, Pages 665–674https://doi.org/10.1145/3097983.3098052Deep autoencoders, and other deep neural networks, have demonstrated their effectiveness in discovering non-linear features across many problem domains. However, in many real-world problems, large outliers and pervasive noise are commonplace, and one ...
- research-articleAugust 2017
Interpretable Predictions of Tree-based Ensembles via Actionable Feature Tweaking
KDD '17: Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data MiningAugust 2017, Pages 465–474https://doi.org/10.1145/3097983.3098039Machine-learned models are often described as "black boxes". In many real-world applications however, models may have to sacrifice predictive power in favour of human-interpretability. When this is the case, feature engineering becomes a crucial task, ...
- research-articleAugust 2017
Multi-Aspect Streaming Tensor Completion
KDD '17: Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data MiningAugust 2017, Pages 435–443https://doi.org/10.1145/3097983.3098007Tensor completion has become an effective computational tool in many real-world data-driven applications. Beyond traditional static setting, with the increasing popularity of high velocity streaming data, it requires efficient online processing without ...
- research-articleAugust 2017
Coresets for Kernel Regression
KDD '17: Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data MiningAugust 2017, Pages 645–654https://doi.org/10.1145/3097983.3098000Kernel regression is an essential and ubiquitous tool for non-parametric data analysis, particularly popular among time series and spatial data. However, the central operation which is performed many times, evaluating a kernel on the data set, takes ...
- research-articleAugust 2017
Robust Top-k Multiclass SVM for Visual Category Recognition
KDD '17: Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data MiningAugust 2017, Pages 75–83https://doi.org/10.1145/3097983.3097991Classification problems with a large number of classes inevitably involve overlapping or similar classes. In such cases it seems reasonable to allow the learning algorithm to make mistakes on similar classes, as long as the true class is still among the ...
- research-articleAugust 2017
Towards an Optimal Subspace for K-Means
KDD '17: Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data MiningAugust 2017, Pages 365–373https://doi.org/10.1145/3097983.3097989Is there an optimal dimensionality reduction for k-means, revealing the prominent cluster structure hidden in the data? We propose SUBKMEANS, which extends the classic k-means algorithm. The goal of this algorithm is twofold: find a sufficient k-means-...