Reproduce “Deep Learning-Based Perceptual Stimulus Encoder for Bionic Vision” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Lucas Relic, Bowen Zhang, Yi-Lin Tuan, Michael Beyeler
The authors train an encoder to choose implant electrodes by evaluating activations through a patient-specific model of percepts. Because the model is nondifferentiable, they train a surrogate and optimize on MNIST images. In simulation, it achieves lower pixel-wise mean squared error than a direct inverse-model baseline across three subject-specific models, but this does not establish improved vision in users.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.