Joseph Scott
joe.scott.ai@gmail.com · joe-scott.com · github.com/j29scott · linkedin.com/in/joseph-scott-ai
Experience
| 2024 – present | Founder, stealth startup in quantitative finance |
Education
| 2017 – 2024 | Ph.D., Computer Science, University of Waterloo. Advisor: Vijay Ganesh. Thesis: Meta-Solving via Machine Learning for Automated Reasoning |
| 2015 – 2017 | M.Math, Computer Science, University of Waterloo |
| 2011 – 2015 | B.S., Computer Science and B.S., Mathematics, Ohio University |
Interests
Artificial intelligence. Automated reasoning and logic solvers: propositional satisfiability, satisfiability modulo theories, mixed-integer linear programming, neural network verification, algorithm selection and meta-solvers. Machine learning: supervised, semi-supervised, unsupervised, and reinforcement learning; neuro-symbolic AI; meta-learning and AutoML; symbolic regression; active learning; applications of LLMs; AI safety. Applications: software testing, physics and automated scientific discovery, quantitative finance.
Selected Publications
Full list on the research page.
- Algorithm Selection for SMT. Scott, Niemetz, Preiner, Nejati, Ganesh. STTT 25(2), 2023.
- Goose: A Meta-Solver for Deep Neural Network Verification. Scott, Pan, Khalil, Ganesh. SMT Workshop, IJCAR/FLoC 2022.
- MachSMT: A Machine Learning-based Algorithm Selector for SMT Solvers. Scott, Niemetz, Preiner, Nejati, Ganesh. TACAS 2021.
- BanditFuzz: Fuzzing SMT Solvers with Multi-agent Reinforcement Learning. Scott, Sudula, Rehman, Mora, Ganesh. FM 2021.
- BanditFuzz: A Reinforcement-Learning Based Performance Fuzzer for SMT Solvers. Scott, Mora, Ganesh. VSTTE 2020.
Awards
| 2021 | MachSMT invited to the STTT journal special issue |
| 2020 | AAAI 2020 Best Three Minute Thesis; AAAI 2020 Spotlight Student Paper |
| 2020 | Vector Institute Doctoral Research Grant |
| 2018 | Travel scholarship, 50 Years of Complexity Theory: A Celebration of the Work of Stephen Cook |
| 2017 | University of Waterloo International Doctoral Student Award |
| 2015 | University of Waterloo International Masters Student Award; Most Outstanding Senior, Tau Beta Pi |
| 2014 | Distinguished Junior Award, Tau Beta Pi |
Teaching
| 2017 – 2022 | University of Waterloo, Instructional Apprentice: Logic for Computer Science |
| 2015 – 2017 | University of Waterloo, Teaching Assistant: Logic for Computer Science, Logic for Software, Algorithms |
| 2014 – 2015 | Ohio University, Lab Teaching Assistant and University College Learning Community |
Service
- Memberships: SAT Association; ACM (Ohio University chapter president 2013–2015); IEEE.
- Organization: Waterloo Blockchain + Security Workshop; Waterloo ML + Security + Verification Workshop.
- Reviewing: WFVML ‘23, Nature (2023), Web Intelligence ‘23, ICML workshops ‘23.
- Artifact evaluation: TACAS ‘22, PLDI ‘22, CAV ‘22, CAV ‘23.
- External reviewing: ICISS ‘18, CP ‘18, CAV ‘18, Computing Surveys (2019), ICISS ‘19, IJCAI ‘19, LPAR ‘20, FM ‘21, CP ‘21, CAV ‘21, TSE ‘21, NeSy ‘23.