Peephole: Predicting Network Performance Before Training

Abstract

The quest for performant networks has been a significant force that drives the advancements of deep learning in recent years. While rewarding, improving network design has never been an easy journey. The large design space combined with the tremendous cost required for network training poses a major obstacle to this endeavor. In this work, we propose a new approach to this problem, namely, predicting the performance of a network before training, based on its architecture. Specifically, we develop a unified way to encode individual layers into vectors and bring them together to form an integrated description via LSTM. Taking advantage of the recurrent network's strong expressive power, this method can reliably predict the performances of various network architectures. Our empirical studies showed that it not only achieved accurate predictions but also produced consistent rankings across datasets -- a key desideratum in performance prediction.

References in corpus (4)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Neural Architecture Search with Reinforcement Learning
• Going Deeper with Convolutions
• Freeze-Thaw Bayesian Optimization

Cited by in corpus (9)

• A Comprehensive Survey on Hardware-Aware Neural Architecture Search
• Weight-Sharing Neural Architecture Search: A Battle to Shrink the Optimization Gap
• EcoNAS: Finding Proxies for Economical Neural Architecture Search
• Fast Task-Aware Architecture Inference
• NeuNetS: An Automated Synthesis Engine for Neural Network Design
• A Novel Training Protocol for Performance Predictors of Evolutionary Neural Architecture Search Algorithms
• DC-NAS: Divide-and-Conquer Neural Architecture Search
• Data Proxy Generation for Fast and Efficient Neural Architecture Search
• Cascade Bagging for Accuracy Prediction with Few Training Samples