2019/10/07 by Dilin Wang, Wang, Dilin, Meng Li +7 · 2 citations
Computer Science · #Advanced Neural Network Applications #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Machine Learning and Data Classification #Stochastic Gradient Optimization Techniques
paper · pdf · doi:10.48550/arxiv.1910.03103
openalex publication_date 2019/10/07 · openalex created_date 2019/10/18 · openalex updated_date 2026/07/28
Designing energy-efficient networks is of critical importance for enabling state-of-the-art deep learning in mobile and edge settings where the computation and energy budgets are highly limited. Recently, Liu et al. (2019) framed the search of efficient neural architectures into a continuous splitting process: it iteratively splits existing neurons into multiple off-springs to achieve progressive loss minimization, thus finding novel architectures by gradually growing the neural network. However, this method was not specifically tailored for designing energy-efficient networks, and is computationally expensive on large-scale benchmarks. In this work, we substantially improve Liu et al. (2019) in two significant ways: 1) we incorporate the energy cost of splitting different neurons to better guide the splitting process, thereby discovering more energy-efficient network architectures; 2) we substantially speed up the splitting process of Liu et al. (2019), which requires expensive eigen-decomposition, by proposing a highly scalable Rayleigh-quotient stochastic gradient algorithm. Our fast algorithm allows us to reduce the computational cost of splitting to the same level of typical back-propagation updates and enables efficient implementation on GPU. Extensive empirical results show that our method can train highly accurate and energy-efficient networks on challenging datasets such as ImageNet, improving a variety of baselines, including the pruning-based methods and expert-designed architectures.