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-articleJune 2024
Large Language Models on Mobile Devices: Measurements, Analysis, and Insights
EdgeFM '24: Proceedings of the Workshop on Edge and Mobile Foundation ModelsJune 2024, Pages 1–6https://doi.org/10.1145/3662006.3662059Deploying large language models (LLMs) inference into mobile devices is cost-efficient for companies, and well addresses the privacy concern of users. However, the limited computation capacity and memory constraints of mobile devices hinder their ...
- research-articleMay 2024
Mobile Foundation Model as Firmware
- Jinliang Yuan,
- Chen Yang,
- Dongqi Cai,
- Shihe Wang,
- Xin Yuan,
- Zeling Zhang,
- Xiang Li,
- Dingge Zhang,
- Hanzi Mei,
- Xianqing Jia,
- Shangguang Wang,
- Mengwei Xu
ACM MobiCom '24: Proceedings of the 30th Annual International Conference on Mobile Computing and NetworkingMay 2024, Pages 279–295https://doi.org/10.1145/3636534.3649361In the current AI era, mobile devices such as smartphones are tasked with executing a myriad of deep neural networks (DNNs) locally. It presents a complex landscape, as these models are highly fragmented in terms of architecture, operators, and ...
- research-articleApril 2024
FedRDMA: Communication-Efficient Cross-Silo Federated LLM via Chunked RDMA Transmission
EuroMLSys '24: Proceedings of the 4th Workshop on Machine Learning and SystemsApril 2024, Pages 126–133https://doi.org/10.1145/3642970.3655834Communication overhead is a significant bottleneck in federated learning (FL), which has been exaggerated with the increasing size of AI models. In this paper, we propose FedRDMA, a communication-efficient cross-silo FL system that integrates RDMA into ...
Federated Few-Shot Learning for Mobile NLP
ACM MobiCom '23: Proceedings of the 29th Annual International Conference on Mobile Computing and NetworkingOctober 2023, Article No.: 63, Pages 1–17https://doi.org/10.1145/3570361.3613277Natural language processing (NLP) sees rich mobile applications. To support various language understanding tasks, a foundation NLP model is often fine-tuned in a federated, privacy-preserving setting (FL). This process currently relies on at least ...
- research-articleOctober 2023
Efficient Federated Learning for Modern NLP
ACM MobiCom '23: Proceedings of the 29th Annual International Conference on Mobile Computing and NetworkingOctober 2023, Article No.: 37, Pages 1–16https://doi.org/10.1145/3570361.3592505Transformer-based pre-trained models have revolutionized NLP for superior performance and generality. Fine-tuning pre-trained models for downstream tasks often requires private data, for which federated learning is the de-facto approach (i.e., FedNLP)...
- posterSeptember 2023
FedAdapter: Efficient Federated Learning for Mobile NLP
ACM TURC '23: Proceedings of the ACM Turing Award Celebration Conference - China 2023July 2023, Pages 27–28https://doi.org/10.1145/3603165.3607380Fine-tuning pre-trained models for downstream tasks often requires private data, for which federated learning is the de-facto approach (i.e., FedNLP). However, FedNLP is prohibitively slow due to the large model sizes and the resultant high network/...
- research-articleJuly 2023
Ske2Grid: skeleton-to-grid representation learning for action recognition
ICML'23: Proceedings of the 40th International Conference on Machine LearningJuly 2023, Article No.: 139, Pages 3431–3441This paper presents Ske2Grid, a new representation learning framework for improved skeleton-based action recognition. In Ske2Grid, we define a regular convolution operation upon a novel grid representation of human skeleton, which is a compact image-like ...
- research-articleMay 2023
Towards Practical Few-shot Federated NLP
EuroMLSys '23: Proceedings of the 3rd Workshop on Machine Learning and SystemsMay 2023, Pages 42–48https://doi.org/10.1145/3578356.3592575Transformer-based pre-trained models have emerged as the predominant solution for natural language processing (NLP). Fine-tuning such pre-trained models for downstream tasks often requires a considerable amount of labeled private data. In practice, ...
- research-articleJune 2024
Dynamic normalization and relay for video action recognition
NIPS '21: Proceedings of the 35th International Conference on Neural Information Processing SystemsDecember 2021, Article No.: 843, Pages 11026–11040Convolutional Neural Networks (CNNs) have been the dominant model for video action recognition. Due to the huge memory and compute demand, popular action recognition networks need to be trained with small batch sizes, which makes learning discriminative ...
- short-paperJune 2021
Towards Ubiquitous Learning: A First Measurement of On-Device Training Performance
EMDL'21: Proceedings of the 5th International Workshop on Embedded and Mobile Deep LearningJune 2021, Pages 31–36https://doi.org/10.1145/3469116.3470009We are witnessing the emergence of ubiquitous learning, where each device (smartphones, wearables, IoTs, etc) can learn from their environments either alone or collaboratively. Such a new paradigm is enabled by deep learning techniques, or more ...
- research-articleNovember 2017
Learning supervised scoring ensemble for emotion recognition in the wild
ICMI '17: Proceedings of the 19th ACM International Conference on Multimodal InteractionNovember 2017, Pages 553–560https://doi.org/10.1145/3136755.3143009State-of-the-art approaches for the previous emotion recognition in the wild challenges are usually built on prevailing Convolutional Neural Networks (CNNs). Although there is clear evidence that CNNs with increased depth or width can usually bring ...
- short-paperOctober 2016
HoloNet: towards robust emotion recognition in the wild
ICMI '16: Proceedings of the 18th ACM International Conference on Multimodal InteractionOctober 2016, Pages 472–478https://doi.org/10.1145/2993148.2997639In this paper, we present HoloNet, a well-designed Convolutional Neural Network (CNN) architecture regarding our submissions to the video based sub-challenge of the Emotion Recognition in the Wild (EmotiW) 2016 challenge. In contrast to previous ...
- ArticleJanuary 2016
Adaptive Synopsis of Non-Human Primates' Surveillance Video Based on Behavior Classification
MMM 2016: Proceedings, Part I, of the 22nd International Conference on MultiMedia Modeling - Volume 9516January 2016, Pages 710–721https://doi.org/10.1007/978-3-319-27671-7_60Non-human primates NHPs play a critical role in biomedical research. Automated monitoring and analysis of NHP's behaviors through the surveillance video can greatly support the NHP-related studies. However, little research work has been undertaken yet. ...
- research-articleSeptember 2015
Deep CCA based super vector for action recognition
2015 IEEE International Conference on Image Processing (ICIP)Pages 1945–1949https://doi.org/10.1109/ICIP.2015.7351140Super vector based feature encoding methods have recently produced state-of-the-art performance in video based action recognition. Inspired by the idea of multi-view super vector (MVSV), we propose a novel global representation, deep canonical correlation ...