3 citations · 8 across the 12 of their papers we have counts for
7 papers · 1 filter
Hyperbolic Binary Neural Network
Jun Chen, Jingyang Xiang, Tianxin Huang +2
Binary Neural Network (BNN) converts full-precision weights and activations into their extreme 1-bit counterparts, making it particularly suitable for deployment on lightweight mob…
Unified Data-Free Compression: Pruning and Quantization without Fine-Tuning
Shipeng Bai, Jun Chen, Xintian Shen +2
Structured pruning and quantization are promising approaches for reducing the inference time and memory footprint of neural networks. However, most existing methods require the ori…
Data-Free Quantization via Mixed-Precision Compensation without Fine-Tuning
Jun Chen, Shipeng Bai, Tianxin Huang +3
Neural network quantization is a very promising solution in the field of model compression, but its resulting accuracy highly depends on a training/fine-tuning process and requires…
Learning Discretized Neural Networks under Ricci Flow
Jun Chen, Hanwen Chen, Mengmeng Wang +3
In this paper, we study Discretized Neural Networks (DNNs) composed of low-precision weights and activations, which suffer from either infinite or zero gradients due to the non-dif…
M22: A Communication-Efficient Algorithm for Federated Learning Inspired by Rate-Distortion
Yangyi Liu, Stefano Rini, Sadaf Salehkalaibar +1
In federated learning (FL), the communication constraint between the remote learners and the Parameter Server (PS) is a crucial bottleneck. For this reason, model updates must be c…
Dynamically Stable Poincaré Embeddings for Neural Manifolds
Jun Chen, Yuang Liu, Xiangrui Zhao +2
In a Riemannian manifold, the Ricci flow is a partial differential equation for evolving the metric to become more regular. We hope that topological structures from such metrics ma…
Thoughts on the Consistency between Ricci Flow and Neural Network Behavior
Jun Chen, Tianxin Huang, Wenzhou Chen +1
The Ricci flow is a partial differential equation for evolving the metric in a Riemannian manifold to make it more regular. On the other hand, neural networks seem to have similar…