AnytimeNet: Controlling Time-Quality TradeOffs in Deep Neural Network Architectures
Last modified: Tue May 12 15:03:39 2020 GMT.
Deeper neural networks, especially those with extremely large numbers of internal parameters, impose a heavy computational burden in obtaining sufficiently high-quality results. These burdens are impeding the application of machine learning and related techniques to time-critical computing systems. To address this challenge, we are proposing an architectural approach for neural networks that adaptively trades off computation time and solution quality to achieve high-quality solutions with timeliness. We propose a novel and general framework, AnytimeNet, that gradually inserts additional layers, so users can expect monotonically increasing quality of solutions as more computation time is expended. The framework allows users to select on the fly when to retrieve a result during runtime. Extensive evaluation results on classification tasks demonstrate that our proposed architecture provides adaptive control of classification solution quality according to the available computation time.
PublishedIn Proceedings of the 2020 Design, Automation, and Test in Europe Conference & Exhibition (DATE'20), Grenoble, France, May 2020.
Copyright © 1996-2023 The FLINT Group
<flint at cs dot yale dot edu>
Yale University Department of Computer Science