cd/entity/Mark Horowitz· home› entities› Mark Horowitz
grep -l @mark horowitz /news/*.json | wc -l → 2

Mark Horowitz

mentions 2 type Person feed RSS

// recent coverage 2 mentions

23:04
2026-06-18
devclubhouse.com
neural-networks

Demystifying Integer Quantization for Neural Network Inference

Integer quantization reduces neural network memory and energy costs by converting high-precision values to lower-bit integers, with INT8 additions consuming 30 times less energy than FP32. The techniq…

19:25
2026-06-18
hello-fri-end.github.io
machine-learning

Integer Quantization: Deep Dive

Integer quantization reduces memory and energy consumption in large language models by representing weights and activations with fewer bits, enabling 70B models to fit on a single GPU in 4-bit precisi…

// co-occurs with top 4 entities
Stanford University 2 INT8 1 FP32 1 LLM 1
// topics top 6 topics
machine learning 2 ai infrastructure 2 large language models 2 neural networks 1 ai chips 1 ai research 1