{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/random-feedback-weights-support-learning-in","title":"Random feedback weights support learning in deep neural networks","arxiv_id":"1411.0247","date":"2014-11-02","proceeding":null,"authors":["Timothy P. Lillicrap","Daniel Cownden","Douglas B. Tweed","Colin J. Akerman"],"abstract":"The brain processes information through many layers of neurons. This deep\narchitecture is representationally powerful, but it complicates learning by\nmaking it hard to identify the responsible neurons when a mistake is made. In\nmachine learning, the backpropagation algorithm assigns blame to a neuron by\ncomputing exactly how it contributed to an error. To do this, it multiplies\nerror signals by matrices consisting of all the synaptic weights on the\nneuron's axon and farther downstream. This operation requires a precisely\nchoreographed transport of synaptic weight information, which is thought to be\nimpossible in the brain. Here we present a surprisingly simple algorithm for\ndeep learning, which assigns blame by multiplying error signals by random\nsynaptic weights. We show that a network can learn to extract useful\ninformation from signals sent through these random feedback connections. In\nessence, the network learns to learn. We demonstrate that this new mechanism\nperforms as quickly and accurately as backpropagation on a variety of problems\nand describe the principles which underlie its function. Our demonstration\nprovides a plausible basis for how a neuron can be adapted using error signals\ngenerated at distal locations in the brain, and thus dispels long-held\nassumptions about the algorithmic constraints on learning in neural circuits.","url_abs":"http://arxiv.org/abs/1411.0247v1","url_pdf":"http://arxiv.org/pdf/1411.0247v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"random-feedback-weights-support-learning-in","repo_url":"https://github.com/jsalbert/biotorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[],"methods":[{"method_slug":"dfa","method_name":"FA"}],"datasets_introduced":[],"methods_introduced":[{"slug":"dfa","name":"FA","full_name":"Feedback Alignment"}],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1411.0247","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}