Per-Tensor Fixed-Point Quantization of the Back-Propagation Algorithm

Authors: Charbel Sakr, Naresh Shanbhag

ICLR 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We demonstrate FX training on three deep learning benchmarks (CIFAR-10, CIFAR-100, SVHN) achieving high fidelity to our FL baseline in that we observe no loss of accuracy higher then 0.56% in all of our experiments. Our precision assignment is further shown to be within 1-b per-tensor of the minimum. We show that our precision assignment methodology reduces representational, computational, and communication costs of training by up to 6 , 8 , and 4 , respectively, compared to the FL baseline and related works. Section 4 is titled "NUMERICAL RESULTS".
Researcher Affiliation Academia Charbel Sakr & Naresh Shanbhag Department of Electrical and Computer Engineering University of Illinois at Urbana-Champaign Illinois, IL 61801, USA {sakr2,shanbhag}@illinois.edu
Pseudocode No The paper describes its methodology in text and mathematical formulas but does not include any clearly labeled pseudocode or algorithm blocks.
Open Source Code No The paper does not contain any explicit statements about releasing source code or provide links to a code repository for the described methodology.
Open Datasets Yes We employ three deep learning benchmarking datasets: CIFAR-10, CIFAR-100 (Krizhevsky and Hinton, 2009), and SVHN (Netzer et al., 2011).
Dataset Splits No Appendix E states, "The value of B(min) is swept and pm i evaluated on the validation set." However, the paper does not specify exact train/validation/test split percentages or sample counts, nor does it reference predefined splits for these datasets.
Hardware Specification Yes All experiments were done using a Pascal P100 NVIDIA GPU.
Software Dependencies No The paper mentions general computing environments like GPUs and CPUs and concepts like 32-bit floating-point arithmetic, but does not provide specific software names with version numbers (e.g., PyTorch 1.9, TensorFlow 2.x).
Experiment Setup Yes The precision configuration Co, with target pm 1%, β0 5%, and η0 1%, via our proposed method is depicted in Figure 2 for each of the four networks considered. ... The mini-batch size we used in all our experiments was 256. ... The smallest value of B(min) resulting in pm < 1% is equal to 4 bits. ... The smallest learning rate value used in the training, which in our case is 0.0001.