Gaussian RAM: Lightweight image classification via stochastic retina-inspired glimpse and reinforcement learning

D Shim, HJ Kim - … on Control, Automation and Systems (ICCAS), 2020 - ieeexplore.ieee.org
Previous studies on image classification have mainly focused on the performance of the
networks, not on real-time operation or model compression. We propose a Gaussian Deep …

Gaussian RAM: Lightweight Image Classification via Stochastic Retina-Inspired Glimpse and Reinforcement Learning

D Shim, HJ Kim - 2020 20th International Conference on Control …, 2020 - dl.acm.org
Previous studies on image classification have mainly focused on the performance of the
networks, not on real-time operation or model compression. We propose a Gaussian Deep …

Gaussian RAM: Lightweight Image Classification via Stochastic Retina-Inspired Glimpse and Reinforcement Learning

D Shim, HJ Kim - 제어로봇시스템학회국제학술대회논문집, 2020 - dbpia.co.kr
Previous studies on image classification have mainly focused on the performance of the
networks, not on real-time operation or model compression. We propose a Gaussian Deep …

Gaussian RAM: Lightweight Image Classification via Stochastic Retina-Inspired Glimpse and Reinforcement Learning

D Shim, HJ Kim - arXiv preprint arXiv:2011.06190, 2020 - arxiv.org
Previous studies on image classification have mainly focused on the performance of the
networks, not on real-time operation or model compression. We propose a Gaussian Deep …

Gaussian RAM: Lightweight Image Classification via Stochastic Retina-Inspired Glimpse and Reinforcement Learning

D Shim, HJ Kim - arXiv e-prints, 2020 - ui.adsabs.harvard.edu
Previous studies on image classification have mainly focused on the performance of the
networks, not on real-time operation or model compression. We propose a Gaussian Deep …

[PDF][PDF] Gaussian RAM: Lightweight Image Classification via Stochastic Retina-Inspired Glimpse and Reinforcement Learning

D Shim, HJ Kim - researchgate.net
Previous studies on image classification have mainly focused on the performance of the
networks, not on real-time operation or model compression. We propose a Gaussian Deep …