Binary stochastic forward-forward training reaches near-real-valued forward-forward accuracy on image benchmarks while estimating 10-100x energy savings in p-bit hardware.
Probabilistic Binary Neural Networks
1 Pith paper cite this work. Polarity classification is still indexing.
abstract
Low bit-width weights and activations are an effective way of combating the increasing need for both memory and compute power of Deep Neural Networks. In this work, we present a probabilistic training method for Neural Network with both binary weights and activations, called BLRNet. By embracing stochasticity during training, we circumvent the need to approximate the gradient of non-differentiable functions such as sign(), while still obtaining a fully Binary Neural Network at test time. Moreover, it allows for anytime ensemble predictions for improved performance and uncertainty estimates by sampling from the weight distribution. Since all operations in a layer of the BLRNet operate on random variables, we introduce stochastic versions of Batch Normalization and max pooling, which transfer well to a deterministic network at test time. We evaluate the BLRNet on multiple standardized benchmarks.
citation-role summary
citation-polarity summary
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
support 1representative citing papers
citing papers explorer
-
Energy-Efficient Supervised Learning with a Binary Stochastic Forward-Forward Algorithm
Binary stochastic forward-forward training reaches near-real-valued forward-forward accuracy on image benchmarks while estimating 10-100x energy savings in p-bit hardware.