{"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/reluplex-an-efficient-smt-solver-for","title":"Reluplex: An Efficient SMT Solver for Verifying Deep Neural Networks","arxiv_id":"1702.01135","date":"2017-02-03","proceeding":null,"authors":["Guy Katz","Clark Barrett","David Dill","Kyle Julian","Mykel Kochenderfer"],"abstract":"Deep neural networks have emerged as a widely used and effective means for\ntackling complex, real-world problems. However, a major obstacle in applying\nthem to safety-critical systems is the great difficulty in providing formal\nguarantees about their behavior. We present a novel, scalable, and efficient\ntechnique for verifying properties of deep neural networks (or providing\ncounter-examples). The technique is based on the simplex method, extended to\nhandle the non-convex Rectified Linear Unit (ReLU) activation function, which\nis a crucial ingredient in many modern neural networks. The verification\nprocedure tackles neural networks as a whole, without making any simplifying\nassumptions. We evaluated our technique on a prototype deep neural network\nimplementation of the next-generation airborne collision avoidance system for\nunmanned aircraft (ACAS Xu). Results show that our technique can successfully\nprove properties of networks that are an order of magnitude larger than the\nlargest networks verified using existing methods.","url_abs":"http://arxiv.org/abs/1702.01135v2","url_pdf":"http://arxiv.org/pdf/1702.01135v2.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":"reluplex-an-efficient-smt-solver-for","repo_url":"https://github.com/guykatzz/ReluplexCav2017","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"reluplex-an-efficient-smt-solver-for","repo_url":"https://github.com/Lzsxx/Leaky-Reluplex","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"reluplex-an-efficient-smt-solver-for","repo_url":"https://github.com/eth-sri/eran","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"reluplex-an-efficient-smt-solver-for","repo_url":"https://github.com/hypro/hypro","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"reluplex-an-efficient-smt-solver-for","repo_url":"https://github.com/pauls658/ReluDiff-ICSE2020-Artifact","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"reluplex-an-efficient-smt-solver-for","repo_url":"https://github.com/rm2pt/veriprune","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"reluplex-an-efficient-smt-solver-for","repo_url":"https://github.com/sen-uni-kn/specrepair","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"reluplex-an-efficient-smt-solver-for","repo_url":"https://github.com/stanleybak/nnenum","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"GPL-3.0"}},{"paper_slug":"reluplex-an-efficient-smt-solver-for","repo_url":"https://github.com/vehicle-lang/vehicle","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"NOASSERTION"}}],"tasks":[{"task_slug":"collision-avoidance","task_name":"Collision Avoidance"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1702.01135","atlas_url":"https://app.syntology.ai/?focus=1702.01135","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}