Lydia Chen has papers accepted and appearing on flagship conferences and journals in various areas of machine learning systems

Machine Learning and Artificial Intelligent Special Issues of IEEE Transactions on Parallel and Distributed Systems

• SGD_Tucker: A Novel Stochastic Optimization Strategy for Parallel Sparse Tucker Decomposition (link)
• Accelerating Gossip-based Deep Learning in Heterogeneous Edge Computing Platforms (link)

Deep Learning with Dirty Label Problems

• ''Online Label Aggregation: A Variational Bayesian Approach'' will appear in WWW 21 (link)

Queuing Modeling for replication systems

• ''Practical Analysis of Replication-Based Systems'' will appear in INFOCOM 21

Deep Models Edge inference

• ''MASA: Responsive Multi-DNN Inference on the Edge'' will appear in PerCom 21, research track
• ''MEMA: Fast Inference of Multiple Deep Models' will appear in PerCom 21, industry track

Hyper Parameter Tuning for Deep Model Cluster

• ''Courier: Real-Time Optimal Batch Size Prediction for Latency SLOs in BigDL'' will appear in ICPE21, industry track