IEEE

Improving the Post-Training Neural Network Quantization by Prepositive Feature Quantization

Abstract: Post-training neural network quantization (PTQ) is an effective model compression technology that has revolutionized the deployment of deep neural networks on various edge devices. It ...
eLife

A differentiable model for optimizing the genetic drivers of synaptogenesis

This study presents SynaptoGen, a differentiable extension of connectome models that links gene expression, protein-protein interaction probabilities, synaptic multiplicity, and synaptic weights, and ...
IEEE

RefQSR: Reference-Based Quantization for Image Super-Resolution Networks

Abstract: Single image super-resolution (SISR) aims to reconstruct a high-resolution image from its low-resolution observation. Recent deep learning-based SISR models show high performance at the ...
GitHub

floret: fastText + Bloom embeddings for compact, full-coverage vectors with spaCy

With -hashCount 2, each entry is stored as the sum of 2 rows in the internal subwords hash table. floret supports 1-4 hashes per entry in the embeddings table. By storing an entry in the embedding ...