Optimization of Neuroprosthetic Vision via End-to-End Deep Reinforcement Learning

被引:17
作者
Kucukoglu, Burcu [1 ]
Rueckauer, Bodo [1 ]
Ahmad, Nasir [1 ]
van Steveninck, Jaap de Ruyter [1 ]
Guclu, Umut [1 ]
van Gerven, Marcel [1 ]
机构
[1] Radboud Univ Nijmegen, Donders Inst Brain Cognit & Behav, Dept Artificial Intelligence, Heyendaalseweg 135, NL-6525 AJ Nijmegen, Gelderland, Netherlands
关键词
Visual neuroprosthesis; phosphene vision; deep reinforcement learning; end-to-end optimization; ENVIRONMENT;
D O I
10.1142/S0129065722500526
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Visual neuroprostheses are a promising approach to restore basic sight in visually impaired people. A major challenge is to condense the sensory information contained in a complex environment into meaningful stimulation patterns at low spatial and temporal resolution. Previous approaches considered task-agnostic feature extractors such as edge detectors or semantic segmentation, which are likely suboptimal for specific tasks in complex dynamic environments. As an alternative approach, we propose to optimize stimulation patterns by end-to-end training of a feature extractor using deep reinforcement learning agents in virtual environments. We present a task-oriented evaluation framework to compare different stimulus generation mechanisms, such as static edge-based and adaptive end-to-end approaches like the one introduced here. Our experiments in Atari games show that stimulation patterns obtained via task-dependent end-to-end optimized reinforcement learning result in equivalent or improved performance compared to fixed feature extractors on high difficulty levels. These findings signify the relevance of adaptive reinforcement learning for neuroprosthetic vision in complex environments.
引用
收藏
页数:16
相关论文
共 50 条
[41]   New Results in End-to-end Image and Video Compression by Deep Learning [J].
Ozsoy, Gokberk ;
Yilmaz, Melih ;
Kirmemis, Ogun ;
Tekalp, A. Murat .
2020 28TH SIGNAL PROCESSING AND COMMUNICATIONS APPLICATIONS CONFERENCE (SIU), 2020,
[42]   AADNet: An End-to-End Deep Learning Model for Auditory Attention Decoding [J].
Nguyen, Nhan Duc Thanh ;
Phan, Huy ;
Geirnaert, Simon ;
Mikkelsen, Kaare ;
Kidmose, Preben .
IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING, 2025, 33 :2695-2706
[43]   End-to-end deep reinforcement learning and control with multimodal perception for planetary robotic dual peg-in-hole assembly [J].
Li, Boxin ;
Wang, Zhaokui .
ADVANCES IN SPACE RESEARCH, 2024, 74 (11) :5860-5873
[44]   LiDAR-Based End-to-End Active SLAM Using Deep Reinforcement Learning in Large-Scale Environments [J].
Chen, Jiaying ;
Wu, Keyu ;
Hu, Minghui ;
Suganthan, Ponnuthurai Nagaratnam ;
Makur, Anamitra .
IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY, 2024, 73 (10) :14187-14200
[45]   End-to-end Optimization of Fluidic Lenses [J].
Na, Mulun ;
Romero, Hector A. Jimenez ;
Yang, Xinge ;
Klein, Jonathan ;
Michels, Dominik L. ;
Heidrich, Wolfgang .
PROCEEDINGS SIGGRAPH ASIA 2024 CONFERENCE PAPERS, 2024,
[46]   SGLPER: A safe end-to-end autonomous driving decision framework combining deep reinforcement learning and expert demonstrations via prioritized experience replay and the Gipps model☆ [J].
Cui, Jianping ;
Yuan, Liang ;
Xiao, Wendong ;
Ran, Teng ;
He, Li ;
Zhang, Jianbo .
DISPLAYS, 2025, 88
[47]   End-to-end Surface Optimization for Light Control [J].
Sun, Yuou ;
Deng, Bailin ;
Zhang, Juyong .
ACM TRANSACTIONS ON GRAPHICS, 2025, 44 (03)
[48]   End-to-End High-Level Control of Lower-Limb Exoskeleton for Human Performance Augmentation Based on Deep Reinforcement Learning [J].
Zheng, Ranran ;
Yu, Zhiyuan ;
Liu, Hongwei ;
Chen, Jing ;
Zhao, Zhe ;
Jia, Longfei .
IEEE ACCESS, 2023, 11 :102340-102351
[49]   SEAE: Stable end-to-end autonomous driving using event-triggered attention and exploration-driven deep reinforcement learning [J].
Cui, Jianping ;
Yuan, Liang ;
Xiao, Wendong ;
Ran, Teng ;
He, Li ;
Zhang, Jianbo .
DISPLAYS, 2025, 87
[50]   An End-to-End Learning Framework for Video Compression [J].
Lu, Guo ;
Zhang, Xiaoyun ;
Ouyang, Wanli ;
Chen, Li ;
Gao, Zhiyong ;
Xu, Dong .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2021, 43 (10) :3292-3308