AI Paper Atlas

Research map / Architectures and efficiency / Quantization

The Synthesis of XNOR Recurrent Neural Networks with Stochastic Logic

Arash Ardakani, Zhengyun Ji, Amir Ardakani, Warren Gross

NeurIPS 2019 · Architectures and efficiency › Quantization · approach: Binary neural networks · 10 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
binarize state computations with stochastic computing (XNOR-LSTM) instead of full-precision state (ablation)ELSTM · PTB character-level LMBPC: 1.47 -> 1.52Hurt
binarize state computations with stochastic computing instead of full-precision state (ablation)ELSTM · PTB word-level LMperplexity: 93.5 -> 95.5Hurt
binarize state computations with stochastic computing instead of full-precision state (ablation)ELSTM · CNN QAerror rate: 40.4 -> 43.8Hurt
binarize state computations with stochastic computing instead of full-precision state (ablation)ELSTM · PTB character-level LMcost (XNOR count): 604,000 -> 7,000Improved
use XNOR-LSTM instead of HitNetquantized LSTM · PTB word-level LMperplexity: 110.3 -> 95.5Improved
use XNOR-LSTM instead of LABquantized LSTM · PTB character-level LMBPC: 1.56 -> 1.52Improved
replace IStanh with weighted-state FSM function WIStanh (ablation)FSM approximation of tanh and sigmoid · PTB neuron outputsapproximation of tanh/sigmoidImproved
use XNOR-LSTM instead of non-stochastic binarized LSTM on FPGAhardware implementation · Virtex-7, 300 neuronsarea and throughput: 66K vs 1.1M slices; 3.2 vs 1.8 TOPSImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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