cs.CVNov 7, 2026

Sensitivity-Gated Accumulator Caching for On-Device Learning of Low-Bit Hand Networks

Authors: Soroor Ghandali, Kanika Mamgain, Manjot Kaur, Akanksha Shukla, Shashira Guntuka

Organizations: Department of Applied Data Science, San José State University, San José, CA, USA

Abstract

Small hand-tracking networks are deployed as 4- and 8-bit codes, and they should keep adapting to a user's camera after deployment, on the device and within roughly the memory of the model itself. Quantization-aware training cannot do this cheaply, because it keeps a 32-bit master copy of every weight together with its optimizer state, about 96 bits per weight. Training the codes directly with stochastic rounding needs no copy but injects rounding noise at every step, and the methods that remove this noise keep a residual buffer for every weight. We introduce sensitivity-gated accumulator caching (SGAC), which keeps residual state for only a small and changing set of weight groups. A set-associative cache of 8-bit accumulators collects the sub-step updates of the groups with the largest Fisher-weighted squared code step and writes them into the codes as whole steps; an evicted line is flushed by stochastic rounding, and a group without a line does not move until it is admitted. We prove that the write-back is unbiased and show, under a local quadratic model, that rounding a sub-step update costs loss of the same order as the update's own gain, so that rounding pays only where the expected gradient is large relative to the curvature. On a 383k-parameter hand-keypoint network, SGAC trains with 8.1 bits of state per weight. Scored on the 4/8-bit network that the device runs after adaptation, it has the highest mean accuracy of nine deployed low-bit networks on two simulated cameras. On twelve ImageNet-C corruptions it has the highest average of corrupted and clean accuracy of seven deployed methods, ahead of the PEQA-style, ECO-style and direct low-bit baselines with confidence intervals that exclude zero, and it matches an 8-bit accumulator kept for every weight with 56% of that state.

Figures & tables

Appendix figures & tables1 asset

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. AAAC: Activation-Aware Adaptive Codebooks for 4-bit LLM Weight Quantization

    May 9, 2026Beshr IslamBouli, David JinLLM Quantization4-Bit Quantization

  2. Threshold-Based Early Stopping of Accumulations in Neural Networks with Binary Activation

    Aug 6, 2026Quentin Luquet de Saint-Germain, Massil Ait Abdeslam, Jean Pierre DavidEfficient Neural Network InferenceBinary Neural Networks

  3. Preserving Updates in Low-Bit Recurrent State Quantization

    Oct 11, 2026Haotian Li, Liran Dong, Kaicheng Xiao +1Recurrent-State QuantizationLow-Bit Quantization