{"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/pathnet-evolution-channels-gradient-descent","title":"PathNet: Evolution Channels Gradient Descent in Super Neural Networks","arxiv_id":"1701.08734","date":"2017-01-30","proceeding":null,"authors":["Chrisantha Fernando","Dylan Banarse","Charles Blundell","Yori Zwols","David Ha","Andrei A. Rusu","Alexander Pritzel","Daan Wierstra"],"abstract":"For artificial general intelligence (AGI) it would be efficient if multiple\nusers trained the same giant neural network, permitting parameter reuse,\nwithout catastrophic forgetting. PathNet is a first step in this direction. It\nis a neural network algorithm that uses agents embedded in the neural network\nwhose task is to discover which parts of the network to re-use for new tasks.\nAgents are pathways (views) through the network which determine the subset of\nparameters that are used and updated by the forwards and backwards passes of\nthe backpropogation algorithm. During learning, a tournament selection genetic\nalgorithm is used to select pathways through the neural network for replication\nand mutation. Pathway fitness is the performance of that pathway measured\naccording to a cost function. We demonstrate successful transfer learning;\nfixing the parameters along a path learned on task A and re-evolving a new\npopulation of paths for task B, allows task B to be learned faster than it\ncould be learned from scratch or after fine-tuning. Paths evolved on task B\nre-use parts of the optimal path evolved on task A. Positive transfer was\ndemonstrated for binary MNIST, CIFAR, and SVHN supervised learning\nclassification tasks, and a set of Atari and Labyrinth reinforcement learning\ntasks, suggesting PathNets have general applicability for neural network\ntraining. Finally, PathNet also significantly improves the robustness to\nhyperparameter choices of a parallel asynchronous reinforcement learning\nalgorithm (A3C).","url_abs":"http://arxiv.org/abs/1701.08734v1","url_pdf":"http://arxiv.org/pdf/1701.08734v1.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":"pathnet-evolution-channels-gradient-descent","repo_url":"https://github.com/kimhc6028/pathnet-pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"BSD-3-Clause"}}],"tasks":[{"task_slug":"continual-learning","task_name":"Continual Learning"},{"task_slug":"reinforcement-learning","task_name":"Reinforcement Learning"},{"task_slug":"reinforcement-learning-1","task_name":"Reinforcement Learning (RL)"},{"task_slug":"transfer-learning","task_name":"Transfer Learning"},{"task_slug":"reinforcement-learning-2","task_name":"reinforcement-learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/continual-learning-on-f-celeba-10-tasks","task":"Continual Learning","dataset":"F-CelebA (10 tasks)","model":"PathNet","rank_in_archive_order":5,"of":7,"metrics":{"Acc":"0.5764"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/1701.08734","atlas_url":"https://app.syntology.ai/?focus=1701.08734","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1701.08734"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/kimhc6028/pathnet-pytorch","reach":{"status":"ok","spdx":"BSD-3-Clause"}}],"summary":{"unverified":2},"by_repo_kind":{"listed":{"samples":2,"ran":0,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"58c9d3cbe8296680","entry":"normalize","repo":"kimhc6028/pathnet-pytorch","repo_kind":"listed","path":"common.py","file_url":"https://github.com/kimhc6028/pathnet-pytorch/blob/HEAD/common.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"58c9d3cbe8296680"}},{"code_sha256_prefix":"4a9a2b3bda0d5733","entry":"salt_and_pepper","repo":"kimhc6028/pathnet-pytorch","repo_kind":"listed","path":"common.py","file_url":"https://github.com/kimhc6028/pathnet-pytorch/blob/HEAD/common.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"4a9a2b3bda0d5733"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}