Don't Mess with Backprop: Doubts about Biologically Plausible Deep Learning | Eric Jang

Traducción a Español

Traducción a Español Biologically Plausible Deep Learning (BPDL) is an active research field at the intersection of Neuroscience and Machine Learning, studying how we can train deep neural networks with a "learning rule" that could conceivably be implemented in the brain. The line of reasoning that typically motivates BPDL is as follows: A Deep Neural Network (DNN) can learn to perform perception tasks that biological brains are capable of (such as detecting and recognizing objects). If activation units and their weights are to DNNs as what neurons and synapses are to biological brains,…

saved by

• SUJASH AGRAWAL

related reading

• Eric Jang: Don't Mess with Backprop: Doubts about Biologically Plausible Deep Learningblog.evjang.com
• Yes you should understand backprop | by Andrej Karpathy | Mediumkarpathy.medium.com
• Neural networks and deep learningneuralnetworksanddeeplearning.com
• Artificial Neural Nets Finally Yield Clues to How Brains Learn | Quanta Magazinequantamagazine.org
• A Recipe for Training Neural Networkskarpathy.github.io
• Backpropagation and the brain | Nature Reviews Neurosciencenature.com
• Neural networks and deep learningneuralnetworksanddeeplearning.com
• Analogies between Biology and Deep Learning [rough note] -- colah's blogcolah.github.io
• Bayesian Neural Networkscs.toronto.edu
• So how does learning work?cs.mcgill.ca
• [2604.21691] There Will Be a Scientific Theory of Deep Learningarxiv.org
• Neural Networks, Types, and Functional Programming -- colah's blogcolah.github.io