Training Quantized Neural Networks to Global Optimality via Semidefinite Programming

Autor:	Bartan, Burak, Pilanci, Mert
Rok vydání:	2021
Předmět:	Computer Science - Machine Learning Statistics - Machine Learning
Druh dokumentu:	Working Paper
Popis:	Neural networks (NNs) have been extremely successful across many tasks in machine learning. Quantization of NN weights has become an important topic due to its impact on their energy efficiency, inference time and deployment on hardware. Although post-training quantization is well-studied, training optimal quantized NNs involves combinatorial non-convex optimization problems which appear intractable. In this work, we introduce a convex optimization strategy to train quantized NNs with polynomial activations. Our method leverages hidden convexity in two-layer neural networks from the recent literature, semidefinite lifting, and Grothendieck's identity. Surprisingly, we show that certain quantized NN problems can be solved to global optimality in polynomial-time in all relevant parameters via semidefinite relaxations. We present numerical examples to illustrate the effectiveness of our method. Comment: v2: Minor edits in the text. The results are unchanged
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/2105.01420 Zobrazit plný text záznamu View this record from Arxiv