Bob Pepin
Postdoc
Publications
CoDeQ: End-to-End Joint Model Compression with Dead-Zone Quantizer for High-Sparsity and Low-Precision Networks
Arxiv, 2025
Oscillations Make Neural Networks Robust to Quantization
Transactions on Machine Learning Research (TMLR), 2025
PePR: Performance Per Resource Unit as a Metric to Promote Small-scale Deep Learning for Medical Image Analysis
Northern Lights Deep Learning Conference (NLDL 2025), 2025
When Can Memorization Improve Fairness?
Arxiv, 2024