activity
20182020
most citedAdversarial Feature Alignment: Avoid Catastrophic Forgetting in Incremental Task Lifelong Learning

27 citations · 27 across the 1 of their papers we have counts for

collaborators

6 papers

cs.LG2020

Continual Local Training for Better Initialization of Federated Models

Xin Yao, Lifeng Sun

Federated learning (FL) refers to the learning paradigm that trains machine learning models directly in the decentralized systems consisting of smart edge devices without transmitt…

cs.LG201927 cited

Adversarial Feature Alignment: Avoid Catastrophic Forgetting in Incremental Task Lifelong Learning

Xin Yao, Tianchi Huang, Chenglei Wu +2

Human beings are able to master a variety of knowledge and skills with ongoing learning. By contrast, dramatic performance degradation is observed when new tasks are added to an ex…

cs.LG2019

Federated Learning with Unbiased Gradient Aggregation and Controllable Meta Updating

Xin Yao, Tianchi Huang, Rui-Xiao Zhang +2

Federated learning (FL) aims to train machine learning models in the decentralized system consisting of an enormous amount of smart edge devices. Federated averaging (FedAvg), the…

cs.LG2019

Federated Learning with Additional Mechanisms on Clients to Reduce Communication Costs

Xin Yao, Tianchi Huang, Chenglei Wu +2

Federated learning (FL) enables on-device training over distributed networks consisting of a massive amount of modern smart devices, such as smartphones and IoT (Internet of Things…

cs.MM2019

Comyco: Quality-Aware Adaptive Video Streaming via Imitation Learning

Tianchi Huang, Chao Zhou, Rui-Xiao Zhang +3

Learning-based Adaptive Bit Rate~(ABR) method, aiming to learn outstanding strategies without any presumptions, has become one of the research hotspots for adaptive streaming. Howe…

cs.MM2018

Tiyuntsong: A Self-Play Reinforcement Learning Approach for ABR Video Streaming

Tianchi Huang, Xin Yao, Chenglei Wu +3

Existing reinforcement learning~(RL)-based adaptive bitrate~(ABR) approaches outperform the previous fixed control rules based methods by improving the Quality of Experience~(QoE)…