• Publications
  • Influence
Advances and Open Problems in Federated Learning
TLDR
Motivated by the explosive growth in FL research, this paper discusses recent advances and presents an extensive collection of open problems and challenges.
More Effective Distributed ML via a Stale Synchronous Parallel Parameter Server
We propose a parameter server system for distributed ML, which follows a Stale Synchronous Parallel (SSP) model of computation that maximizes the time computational workers spend doing useful work on
Base-delta-immediate compression: Practical data compression for on-chip caches
TLDR
There is a need for a simple yet efficient compression technique that can effectively compress common in-cache data patterns, and has minimal effect on cache access latency.
LOCI: fast outlier detection using the local correlation integral
TLDR
Experiments show that LOCI and aLOCI can automatically detect outliers and micro-clusters, without user-required cut-offs, and that they quickly spot both expected and unexpected outliers.
Synopsis diffusion for robust aggregation in sensor networks
TLDR
This paper presents a general framework for achievingantly more accurate and reliable answers by combining energy-efficient multi-path routing schemes with techniques that avoid double-counting, and demonstrates the significant robustness, accuracy, and energy-efficiency improvements of synopsis diffusion over previous approaches.
Ambit: In-Memory Accelerator for Bulk Bitwise Operations Using Commodity DRAM Technology
TLDR
Ambit is proposed, an Accelerator-in-Memory for bulk bitwise operations that largely exploits existing DRAM structure, and hence incurs low cost on top of commodity DRAM designs (1% of DRAM chip area).
Memory consistency and event ordering in scalable shared-memory multiprocessors
TLDR
A new model of memory consistency, called release consistency, that allows for more buffering and pipelining than previously proposed models is introduced and is shown to be equivalent to the sequential consistency model for parallel programs with sufficient synchronization.
PipeDream: generalized pipeline parallelism for DNN training
TLDR
PipeDream is presented, a system that adds inter-batch pipelining to intra-batch parallelism to further improve parallel training throughput, helping to better overlap computation with communication and reduce the amount of communication when possible.
SybilGuard: Defending Against Sybil Attacks via Social Networks
TLDR
This paper presents SybilGuard, a novel protocol for limiting the corruptive influences of sybil attacks, based on the ldquosocial networkrdquo among user identities, where an edge between two identities indicates a human-established trust relationship.
Memory consistency and event ordering in scalable shared-memory multiprocessors
A new model of memory consistency, called release consistency, that allows for more buffering and pipelining than previously proposed models is introduced. A framework for classifying shared accesses
...
...