36 citations · 124 across the 26 of their papers we have counts for
5 papers · 1 filter
TORQUE: Optimizing What (not) to Quantize Before and After Rotation
Ran Ben Basat, Michael Mitzenmacher, Shay Vargaftik
Uniform random rotations are an effective preprocessing step for quantization: they make normalized coordinate distributions approximately Gaussian, enabling the use of codebooks o…
Entropy-Constrained Adaptive Stochastic Quantization
Ran Ben Basat, Yaniv Ben-Itzhak, Michael Mitzenmacher +1
Adaptive stochastic quantization (ASQ) is a recently introduced quantization approach that optimizes the Mean Squared Error (MSE) for a given input while preserving unbiasedness. I…
Quantizing With Randomized Hadamard Transforms: Efficient Heuristic Now Proven
Ran Ben-Basat, William Kuszmaul, Michael Mitzenmacher +2
Uniform random rotations (URRs) are a common preprocessing step in modern quantization approaches used for gradient compression, inference acceleration, KV-cache compression, model…
A Note on TurboQuant and the Earlier DRIVE/EDEN Line of Work
Ran Ben-Basat, Yaniv Ben-Itzhak, Gal Mendelson +3
This note clarifies the relationship between the recent TurboQuant work and the earlier DRIVE (NeurIPS 2021) and EDEN (ICML 2022) schemes. DRIVE is a 1-bit quantizer that EDEN exte…
DynamiQ: Accelerating Gradient Synchronization using Compressed Multi-hop All-reduce
Wenchen Han, Shay Vargaftik, Michael Mitzenmacher +1
Multi-hop all-reduce is the de facto backbone of large model training. As the training scale increases, the network often becomes a bottleneck, motivating the reduction of the volu…