activity
20172022
most citedImproving Neural Network Quantization without Retraining using Outlier Channel Splitting

151 citations · 185 across the 12 of their papers we have counts for

collaborators

15 papers

cs.NI2021

A Roadmap for Enabling a Future-Proof In-Network Computing Data Plane Ecosystem

Daehyeok Kim, Nikita Lazarev, Tommy Tracy +7

As the vision of in-network computing becomes more mature, we see two parallel evolutionary trends. First, we see the evolution of richer, more demanding applications that require…

cs.CV20211 cited

Dense Pruning of Pointwise Convolutions in the Frequency Domain

Mark Buckler, Neil Adit, Yuwei Hu +2

Depthwise separable convolutions and frequency-domain convolutions are two recent ideas for building efficient convolutional neural networks. They are seemingly incompatible: the v…

cs.AR2021

Dagger: Accelerating RPCs in Cloud Microservices Through Tightly-Coupled Reconfigurable NICs

Nikita Lazarev, Shaojie Xiang, Neil Adit +2

The ongoing shift of cloud services from monolithic designs to microservices creates high demand for efficient and high performance datacenter networking stacks, optimized for fine…

cs.AR2021

Enabling Design Methodologies and Future Trends for Edge AI: Specialization and Co-design

Cong Hao, Jordan Dotzel, Jinjun Xiong +3

Artificial intelligence (AI) technologies have dramatically advanced in recent years, resulting in revolutionary changes in people's lives. Empowered by edge computing, AI workload…

cs.LG2021

SPADE: A Spectral Method for Black-Box Adversarial Robustness Evaluation

Wuxinlin Cheng, Chenhui Deng, Zhiqiang Zhao +3

A black-box spectral method is introduced for evaluating the adversarial robustness of a given machine learning (ML) model. Our approach, named SPADE, exploits bijective distance m…

cs.LG20201 cited

FracBNN: Accurate and FPGA-Efficient Binary Neural Networks with Fractional Activations

Yichi Zhang, Junhao Pan, Xinheng Liu +3

Binary neural networks (BNNs) have 1-bit weights and activations. Such networks are well suited for FPGAs, as their dominant computations are bitwise arithmetic and the memory requ…