Skip to the content.

About Me

I am a staff scientist at Latitude AI focusing on improving autonomy for self-driving vehicles. Before that I was a researcher at Microsoft Research in Redmond, WA. I previously obtained a PhD in Computer Science from the University of Texas at Austin advised by Professor Peter Stone.

I work at the intersection of deep learning and reinforcement learning to develop autonomy capable of adapting and learning in complex environments.

Publications

Uni[Mask]: Unified Inference in Sequential Decision Problems
M Carroll, O Paradise, J Lin, R Georgescu, M Sun, D Bignell, S Milani, K Hofmann, M Hausknecht, A Dragan, S Devlin. 2022 Conference on Neural Information Processing Systems (NeurIPS Oral)
MoCapAct: A Multi-Task Dataset for Simulated Humanoid Control
N Wagener, A Kolobov, F Frujeri, R Loynd, C Cheng, M Hausknecht. 2022 Conference on Neural Information Processing Systems: Datasets and Benchmarks Track (NeurIPS)
[Webpage] [Blog Post] [Code]
Consistent Dropout for Policy Gradient Reinforcement Learning
M Hausknecht, N Wagener. 2022 arXiv Preprint
Reading and Acting while Blindfolded: The Need for Semantics in Text Game Agents
S Yao, K Narasimhan, M Hausknecht. 2021 Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL)
[BlogPost]
ALFWorld: Aligning Text and Embodied Environments for Interactive Learning
M Shridhar, X Yuan, M Côté, Y Bisk, A Trischler, M Hausknecht. International Conference on Learning Representations (ICLR) 2021
[Webpage]
Keep CALM and Explore: Language Models for Action Generation in Text-based Games
S Yao, R Rao, M Hausknecht, K Narasimhan. Empirical Methods in Natural Language Processing (EMNLP) 2020
[Code][Video]
Working Memory Graphs
R Loynd, R Fernandez, A Celikyilmaz, A Swaminathan, M Hausknecht. International Conference on Machine Learning (ICML) 2020
[Code][Video]
Learning Calibratable Policies using Programmatic Style-Consistency
E Zhan, A Tseng, Y Yue, A Swaminathan, M Hausknecht. International Conference on Machine Learning (ICML) 2020
Graph Constrained Reinforcement Learning for Natural Language Action Spaces
P Ammanabrolu, M Hausknecht. International Conference on Learning Representations (ICLR) 2020
[Code]
Interactive Fiction Games: A Colossal Adventure
M Hausknecht, P Ammanabrolu, M Côté, X Yuan. Association for the Advancement of Artificial Intelligence (AAAI) 2020
[Code][Blog Post]
Scriptnet: Neural static analysis for malicious javascript detection
J Stokes, R Agrawal, G McDonald, M Hausknecht.IEEE Military Communications Conference (MILCOM) 2019
Nail: A general interactive fiction agent
M Hausknecht, R Loynd, G Yang, A Swaminathan, JD Williams.Technical Report 2019
[Code]
Counting to Explore and Generalize in Text-based Games
X Yuan, M Côté, A Sordoni, R Laroche, R Tachet des Combes, M Hausknecht, A Trischler. European Workshop on Reinforcement Learning (EWRL) 2018
TextWorld: A Learning Environment for Text-based Games
M Côté, A Kadar, X Yuan, B Kybartas, T Barnes, E Fine, J Moore, M Hausknecht, L Asri, M Adada, W Tay, A Trischler. IJCAI/ICML Computer Games Workshop 2018
[Code]
Leveraging grammar and reinforcement learning for neural program synthesis
R Bunel, M Hausknecht, J Devlin, R Singh, P Kohli. International Conference on Learning Representations (ICLR) 2018
Revisiting the arcade learning environment: Evaluation protocols and open problems for general agents
MC Machado, MG Bellemare, E Talvitie, J Veness, M Hausknecht, M Bowling. International Joint Conferences on Artificial Intelligence (IJCAI) 2017
Neural Program Meta-Induction
J Devlin, RR Bunel, R Singh, M Hausknecht, P Kohli.Advances in Neural Information Processing Systems (NeurIPS) 2017
Cooperation and communication in multiagent deep reinforcement learning
Matthew Hausknecht. PhD Thesis 2016
Half field offense: An environment for multiagent learning and ad hoc teamwork
M Hausknecht, P Mupparaju, S Subramanian, S Kalyanakrishnan, P Stone. AAMAS Adaptive Learning Agents (ALA) Workshop 2016
On-policy vs. off-policy updates for deep reinforcement learning
M Hausknecht, P Stone.Deep Reinforcement Learning: Frontiers and Challenges, (IJCAI) Workshop 2016
Deep Reinforcement Learning in Parameterized Action Space
M Hausknecht, P Stone. Proceedings of the International Conference on Learning Representations (ICLR) 2016
[Code]
Machine Learning Capabilities of a Simulated Cerebellum
M Hausknecht, W Li, M Mauk, and P Stone. IEEE Transactions on Neural Networks and Learning Systems (TNNLS) 2016
Deep Recurrent Q-Learning for Partially Observable MDPs
M Hausknecht, P Stone. AAAI Fall Symposium on Sequential Decision Making for Intelligent Agents 2015
Beyond Short Snippets: Deep Networks for Video Classification
J Ng, M Hausknecht, S Vijayanarasimhan, O Vinyals, R Monga, G Toderici. Conference on Computer Vision and Pattern Recognition (CVPR) 2015
A Neuroevolution Approach to General Atari Game Playing
M Hausknecht, J Lehman, R Miikkulainen, and P Stone. IEEE Transactions on Computational Intelligence and AI in Games (TCIAIG) 2014
Using a million cell simulation of the cerebellum: Network scaling and task generality
W Li, M Hausknecht, P Stone, and M Mauk. Neural Networks 2012
HyperNEAT-GGP: A HyperNEAT-based Atari General Game Player
M Hausknecht, P Khandelwal, R Miikkulainen, and P Stone. Proceedings of Genetic and Evolutionary Computation Conference (GECCO) 2012
Dynamic Lane Reversal in Traffic Management
M Hausknecht, T Au, P Stone, D Fajardo, and T Waller. Proceedings of IEEE Intelligent Transportation Systems Conference (ITSC) 2011
Autonomous Intersection Management: Multi-Intersection Optimization
M Hausknecht, T Au, and P Stone. IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) 2011
Learning Powerful Kicks on the Aibo ERS-7: The Quest for a Striker
M Hausknecht, P Stone. Proceedings of the RoboCup International Symposium 2010
For want of a nail: How absences cause events
P Wolff, A Barbey, M Hausknecht. Journal of Experimental Psychology: General 2009

Open Source Software

I’ve built or contributed significantly to the following open-source projects:

Jericho is a lightweight python-based interface connecting learning agents with interactive fiction games. It serves as a testbed for benchmarking progress of language-based agents on man-made text games. BlogPost
Half Field Offense (HFO) is a multiagent subtask in RoboCup simulated soccer, modeling a situation in which the offense of one team has to get past the defense of the opposition in order to shoot goals.
Arcade Learning Environment (ALE) Is a well known platform for training learning agents to play Atari games. I developed the first external interface to ALE which allowed the environment to be used as a library.