Matthew Landers
Home
Experience
Implementations
CV
Notes
About these notes
Glossary
Pseudocode
About RL
MDPs
Value Func. & Policies
DP for MDPs
DP for MDPs Proofs
Model-Free Prediction
Prediction with Approx.
Model-Free Control
On-Policy Control
with Approximation
Off-Policy Control
with Approximation
Importance Sampling
Learnability of
RL Objectives
The Deadly Triad
Deep Q-Learning
Policy Gradients
Actor-Critic Framework
DPG & DDPG
Reparameterization Trick
Soft Actor-Critic
Group Relative Policy
Optimization
Transformer
TRPO
Conjugate Gradient Method
PPO
Double Q-Learning
TD3
NN Verification
DRL Verification
AlphaZero (chess)
Bayes’ for Distributions
Backpropagation
Off-Policy Evaluation
Constrained MDPs
Constrained Policy 
Optimization
PID Lagrangian
Successor Features
Policy Distillation
KL Divergence
Implicit Q-Learning
Cross-Entropy Method
Prior-Data Fitted Network
TabPFN

Matthew Landers

image of Matt Landers 

AI Researcher, Member of Technical Staff at Plastic Labs

CV · Google Scholar · LinkedIn

qwp4pk [at] virginia [dot] edu

I am a Member of Technical Staff at Plastic Labs, where I work on memory and reasoning systems for AI agents.

I earned my Ph.D. in Computer Science from the University of Virginia, advised by Tom Hartvigsen and Afsaneh Doryab, and my M.S. in Computer Science from Johns Hopkins University, advised by Suchi Saria.

In my research, I focus on structured and scalable reinforcement learning for high-dimensional decision-making, with applications to foundation models, agents, and real-world control systems.

Previously, I applied reinforcement learning to production control systems and spent several years as a software engineer and startup founder.

Papers

Improving and Accelerating Offline RL in Large Discrete Action Spaces with Structured Policy Initialization, ICLR (2026)

Matthew Landers, Taylor W. Killian, Thomas Hartvigsen, Afsaneh Doryab

BraVE: Offline Reinforcement Learning for Discrete Combinatorial Action Spaces, NeurIPS (2025)

Matthew Landers, Taylor W. Killian, Hugo Barnes, Thomas Hartvigsen, Afsaneh Doryab

SAINT: Attention-Based Policies for Discrete Combinatorial Action Spaces, Under Review (2025)

Matthew Landers, Taylor W. Killian, Thomas Hartvigsen, Afsaneh Doryab

Parameter Transfer for Single-Task Reinforcement Learning, International Joint Conference on Neural Networks (2025)

Matthew Landers and Afsaneh Doryab

Deep Reinforcement Learning Verification: A Survey, ACM Computing Surveys (2023)

Matthew Landers and Afsaneh Doryab

A bias evaluation checklist for predictive models and its pilot application for 30-day hospital readmission models, Journal of the American Medical Informatics Association (2022)

Echo Wang*, Matthew Landers*, Roy Adams*, Adarsh Subbaswamy, Hadi Kharrazi, Darrell Gaskin, and Suchi Saria

Digital Endpoints: Definition, Benefits, and Current Barriers in Accelerating Development and Adoption, Digital Biomarkers (2021)

Matthew Landers, Ray Dorsey, and Suchi Saria

Will Artificial Intelligence Replace the Movement Disorders Specialist for Diagnosing and Managing Parkinson’s Disease?, Journal of Parkinson’s Disease (2021)

Matthew Landers, Suchi Saria, and Alberto Espay

Prediction of Hospital Readmission from Longitudinal Mobile Data Streams, Sensors (2021)

Chen Qian, Patraporn Leelaprachakul, Matthew Landers, Carissa Low, Anind K . Dey, and Afsaneh Doryab

* indicates equal contribution