Scholar
Garrett Wilson
Google Scholar ID: vYUcuEUAAAAJ
Meta Platforms, Inc.
Citations & Impact
Citations
1,612
H-index
6
i10-index
5
Publications
9
Co-authors
10
list available
Publications
Resume (English only)
Academic Achievements
- [{'Paper': 'Predictive Response Optimization: Using Reinforcement Learning to Fight Online Social Network Abuse', 'Conference/Journal': 'USENIX Security Symposium, 2025'}, {'Paper': 'CALDA: Improving Multi-Source Time Series Domain Adaptation With Contrastive Adversarial Learning', 'Conference/Journal': 'IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023'}, {'Paper': 'Domain Adaptation Under Behavioral and Temporal Shifts for Natural Time Series Mobile Activity Recognition', 'Conference/Journal': '8th SIGKDD International Workshop on Mining and Learning from Time Series, 2022'}, {'Paper': 'Multi-Source Unsupervised Domain Adaptation for Time-Series Sensor Data', 'Conference/Journal': 'Washington State University, 2021'}, {'Paper': 'Multi-Source Deep Domain Adaptation with Weak Supervision for Time-Series Sensor Data', 'Conference/Journal': 'Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, 2020'}, {'Paper': 'A Survey of Unsupervised Deep Domain Adaptation', 'Conference/Journal': 'ACM Transactions on Intelligent Systems and Technology, 2020'}, {'Paper': 'Multi-Purposing Domain Adaptation Discriminators for Pseudo Labeling Confidence', 'Conference/Journal': "AdvML'19: Workshop on Adversarial Learning Methods for Machine Learning and Data Mining at KDD, 2019"}, {'Paper': 'Cyber-physical Support of Daily Activities: A Robot/Smart Home Partnership', 'Conference/Journal': 'ACM Transactions on Cyber-Physical Systems, 2019'}]
Research Experience
- [{'Position': 'Research Scientist', 'Time': '2022 - Present', 'Institution': 'Meta Platforms, Inc.', 'WorkContent': "Built and improved anti-abuse ML systems; Tech Lead for the team's anti-abuse countermetric workstream"}, {'Position': 'Research Assistant', 'Time': '2017 - 2021', 'Institution': 'Washington State University', 'WorkContent': 'Collected datasets and improved model performance for target users with unlabeled data through time series domain adaptation; Helped older adults retain functional independence longer through robotic activity support'}, {'Position': 'Software Engineer Intern', 'Time': 'Summers of 2020, 2021', 'Institution': 'Facebook, Inc.', 'WorkContent': 'Improved ML model performance under domain shifts; Developed PyTorch framework for sharing embeddings across ML models'}, {'Position': 'Software Engineer Intern', 'Time': 'Summers of 2013, 2014, 2015', 'Institution': 'Solution Engineering, Inc.', 'WorkContent': 'Developed and tested a UAV vision landing system prototype; Corrected Raspberry Pi camera driver stride length calculation and Qt Gstreamer YUV to RGB color conversion'}, {'Position': 'STEM Instructor', 'Time': 'Summers of 2016, 2017', 'Institution': 'Sunset Lake Camp', 'WorkContent': "Taught game programming with Unity, Boolean and fuzzy logic with Lego Mindstorms, and radio with a 'fox' hunt; Led students building hydraulic claws, surface-mount software defined radio kits, and through-hole soldering Microbug kits"}, {'Position': 'High School Math & Computer Teacher', 'Time': '2016 - 2017', 'Institution': 'Yap SDA School, Micronesia', 'WorkContent': 'Taught full year of Precalc, Algebra 1 & 2, Geometry, 8th Math, and one semester of 12th Computers; Configured & managed a lab with 18 Windows computers using Windows Deployment Services and batch script unattended software installs'}]
Background
- Interested in machine learning, applied mathematics, and data analysis. For the last 4 years, he has been building and improving ML-based anti-abuse systems for Meta's products.
