AlphaX: eXploring Neural Architectures with Deep Neural Networks and Monte Carlo Tree Search

Abstract

Neural Architecture Search (NAS) has shown great success in automating the design of neural networks, but the prohibitive amount of computations behind current NAS methods requires further investigations in improving the sample efficiency and the network evaluation cost to get better results in a shorter time. In this paper, we present a novel scalable Monte Carlo Tree Search (MCTS) based NAS agent, named AlphaX, to tackle these two aspects. AlphaX improves the search efficiency by adaptively balancing the exploration and exploitation at the state level, and by a Meta-Deep Neural Network (DNN) to predict network accuracies for biasing the search toward a promising region. To amortize the network evaluation cost, AlphaX accelerates MCTS rollouts with a distributed design and reduces the number of epochs in evaluating a network by transfer learning guided with the tree structure in MCTS. In 12 GPU days and 1000 samples, AlphaX found an architecture that reaches 97.84\% top-1 accuracy on CIFAR-10, and 75.5\% top-1 accuracy on ImageNet, exceeding SOTA NAS methods in both the accuracy and sampling efficiency. Particularly, we also evaluate AlphaX on NASBench-101, a large scale NAS dataset; AlphaX is 3x and 2.8x more sample efficient than Random Search and Regularized Evolution in finding the global optimum. Finally, we show the searched architecture improves a variety of vision applications from Neural Style Transfer, to Image Captioning and Object Detection.

another search algorithm for NAS. arXiv admin note: substantial text overlap with arXiv:1805.07440

References in corpus (17)

• Continuous control with deep reinforcement learning
• MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
• Show, Attend and Tell: Neural Image Caption Generation with Visual Attention
• LSTM: A Search Space Odyssey
• Practical Bayesian Optimization of Machine Learning Algorithms
• Neural Architecture Search with Reinforcement Learning
• Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor
• SGDR: Stochastic Gradient Descent with Warm Restarts
• Efficient Neural Architecture Search via Parameter Sharing
• Neural Architecture Optimization
• NAS-Bench-101: Towards Reproducible Neural Architecture Search
• ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design
• Simple And Efficient Architecture Search for Convolutional Neural Networks
• Progressive Neural Architecture Search
• Evaluating the Search Phase of Neural Architecture Search
• DeepArchitect: Automatically Designing and Training Deep Architectures
• A Genetic Programming Approach to Designing Convolutional Neural Network Architectures

Cited by in corpus (27)

• Compute Trends Across Three Eras of Machine Learning
• A Survey on Neural Architecture Search
• Evaluating the Search Phase of Neural Architecture Search
• Neural Predictor based Quantum Architecture Search
• Bridging the Gap between Sample-based and One-shot Neural Architecture Search with BONAS
• ELF OpenGo: An Analysis and Open Reimplementation of AlphaZero
• Sample-Efficient Neural Architecture Search by Learning Action Space
• Learning Search Space Partition for Black-box Optimization using Monte Carlo Tree Search
• Weight-Sharing Neural Architecture Search: A Battle to Shrink the Optimization Gap
• A Generic Graph-based Neural Architecture Encoding Scheme for Predictor-based NAS
• How Powerful are Performance Predictors in Neural Architecture Search?
• Stronger NAS with Weaker Predictors
• Optimal Transport Kernels for Sequential and Parallel Neural Architecture Search
• Designing deep neural networks for driver intention recognition
• Improving One-shot NAS by Suppressing the Posterior Fading
• Few-shot Neural Architecture Search
• CATE: Computation-aware Neural Architecture Encoding with Transformers
• NeuNetS: An Automated Synthesis Engine for Neural Network Design
• Automated Dominative Subspace Mining for Efficient Neural Architecture Search
• Towards Understanding the Effects of Evolving the MCTS UCT Selection Policy
• NAS-HPO-Bench-II: A Benchmark Dataset on Joint Optimization of Convolutional Neural Network Architecture and Training Hyperparameters
• Exploring the Loss Landscape in Neural Architecture Search
• Learn Faster and Forget Slower via Fast and Stable Task Adaptation
• Hyperparameter optimization with REINFORCE and Transformers
• Scaling Up Deep Neural Network Optimization for Edge Inference
• Training BatchNorm Only in Neural Architecture Search and Beyond
• Neural Architecture Search via Combinatorial Multi-Armed Bandit