AI Paper Atlas

Research map / Architectures and efficiency / Quantization

A2Q: Accumulator-Aware Quantization with Guaranteed Overflow Avoidance

Ian Colbert, Alessandro Pappalardo, Jakoba Petri-Koenig

ICCV 2023 · Architectures and efficiency › Quantization · approach: Mixed-precision quantization · 22 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
train with A2Q instead of heuristically reducing data bit widths to hit a target accumulator widthQNNs on CIFAR10 and BSD300 · accumulator-constrained Pareto frontiertask performance vs accumulator bit width: most models show less than a 1% performance drop with a 16-bit accumulatorImproved
reduce accumulator bit width under A2Q l1-norm constraintA2Q QNNs · benchmark averageunstructured sparsity and relative accuracy: up to 92% unstructured weight sparsity while maintaining 99.2% of the floating-point model accuracyImproved
use layer-wise minimized accumulator width instead of fixed 32-bit accumulatorbaseline QAT on FINN · LUT utilization vs task performance Paretoresource-accuracy trade-offImproved
minimize accumulator width from final trained weight l1-norm instead of data type boundbaseline QAT on FINN · LUT utilization vs task performance Paretoresource-accuracy trade-offImproved
train with A2Q rather than baseline QAT with post-hoc accumulator minimizationFINN accelerators · LUT utilization vs task performance Paretoresource-accuracy trade-off: up to 2.3x reduction in resource utilization over 32-bit accumulator counterparts with 99.2% of the floating-point modelImproved
train with A2Q instead of wraparound or saturating accumulators at very low P8-bit binary MNIST linear classifier · binary MNISTtop-1 accuracyImproved
use saturation instead of wraparound accumulation (ablation)8-bit binary MNIST linear classifier · binary MNISTtop-1 accuracyMixed
apply A2Q (round-to-zero) in post-training quantizationA2Q · PTQaccuracyHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy