{"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/generalized-value-iteration-networks-life","title":"Generalized Value Iteration Networks: Life Beyond Lattices","arxiv_id":"1706.02416","date":"2017-06-08","proceeding":null,"authors":["Sufeng. Niu","Siheng Chen","Hanyu Guo","Colin Targonski","Melissa C. Smith","Jelena Kovačević"],"abstract":"In this paper, we introduce a generalized value iteration network (GVIN),\nwhich is an end-to-end neural network planning module. GVIN emulates the value\niteration algorithm by using a novel graph convolution operator, which enables\nGVIN to learn and plan on irregular spatial graphs. We propose three novel\ndifferentiable kernels as graph convolution operators and show that the\nembedding based kernel achieves the best performance. We further propose\nepisodic Q-learning, an improvement upon traditional n-step Q-learning that\nstabilizes training for networks that contain a planning module. Lastly, we\nevaluate GVIN on planning problems in 2D mazes, irregular graphs, and\nreal-world street networks, showing that GVIN generalizes well for both\narbitrary graphs and unseen graphs of larger scale and outperforms a naive\ngeneralization of VIN (discretizing a spatial graph into a 2D image).","url_abs":"http://arxiv.org/abs/1706.02416v2","url_pdf":"http://arxiv.org/pdf/1706.02416v2.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":"generalized-value-iteration-networks-life","repo_url":"https://github.com/sufengniu/GVIN","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"q-learning","task_name":"Q-Learning"}],"methods":[{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"q-learning","method_name":"Q-Learning"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1706.02416","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}