NK
Nathan Kallus
cs.LGstat.MLstat.MEmath.STstat.THmath.OCecon.EMcs.AIcs.IRstat.AP
On Valency
published · living versionsW_ec4xva75·v1 · currentpublished
AT-AT: Automatic Tool Optimization from Agent Traces
with Maya Ravichandran, Pragya Tripathi, Dharamendra Kumar, Adith Swaminathan
1 version
W_bvxjr4hx·v1published
W_fc55ezc8·v1published
Preprints & journals
172 papers in the corpus · 2013–2026Policy Learning with Weak Signals2610.10167v1 · Benedikt Koch, Winston Chou, Aur'elien Bibaut et al.2026 · 0 citationsarXiv
Treatment Effect Modification by Latent Classes2610.10851v1 · Hui Lan, Nathan Kallus, Winston Chou2026 · 0 citationsarXiv
Adaptive inference for functionals of M-estimands2609.39274v1 · James Leiner, Aur'elien Bibaut, Nathan Kallus et al.2026 · 0 citationsarXiv
Scalable Multi-Task Inverse Reinforcement Learning2610.00758v1 · Allen Tran, Jia Wan, Nathan Kallus et al.2026 · 0 citationsarXiv
The future of large-scale experiments and their challenges in the digital era.42754695 · Holtz, David, Bojinov, Iavor, Johari, Ramesh et al.2026 · 0 citationsNature human behaviour. 2026
A positive resolution of the gap-entropy conjecture2609.10529v1 · P. M. Aronow, Nathan Kallus, Patrick Lopatto2026 · 0 citationsarXiv
Bellman Calibration for Marginalized Importance Weighting in Offline Reinforcement Learning2608.24858v1 · Lars van der Laan, Nathan Kallus2026 · 0 citationsarXiv
Recommendation Quality and the Concentration of Consumption: Experimental Evidence from Netflix2608.21274v1 · Guy Aridor, Winston Chou, Nathan Kallus et al.2026 · 0 citationsarXiv
Evaluating for the long term: Learnings from industry2608.08043v1 · Leif Sigerson, Tom Cunningham, Winston Chou et al.2026 · 0 citationsarXiv
A Human-Augmenting Agentic Workflow for Observational Causal Inference2607.22443v1 · Winston Chou, Adrien Alexandre, Lars Olds et al.2026 · 0 citationsarXiv
Semiparametric inference on identification sets in choice modeling2607.21790v1 · Antoine Scheid, Jia Wan, Guy Aridor et al.2026 · 0 citationsarXiv
Fitted Occupancy-Ratio Evaluation without Bellman Completeness2607.05375v2 · Lars van der Laan, Nathan Kallus2026 · 0 citationsarXiv
Mult-DPO: Multinomial Direct Preference Optimization for Recommender Systems2606.10078v1 · Yaochen Zhu, Harald Steck, James McInerney et al.2026 · 0 citationsarXiv
Reward Transfer from Inverse Reinforcement Learning: A Coupled Minimax Approach2605.27834v1 · Guang-Yuan Hao, Lars van der Laan, Aur'elien Bibaut et al.2026 · 0 citationsarXiv
The Context Gathering Decision Process: A POMDP Framework for Agentic Search2605.07042v1 · Chinmaya Kausik, Adith Swaminathan, Nathan Kallus2026 · 0 citationsarXiv
Nonparametric Instrumental Variable Analysis Without Structural Equations: Debiased Inference on Functionals of Inverse Problems with No Solutions2604.24660v2 · Zikai Shen, Nathan Kallus, Dimitri Meunier et al.2026 · 0 citationsarXivon Valency
Functional Natural Policy Gradients2603.28681v2 · Aurelien Bibaut, Houssam Zenati, Thibaud Rahier et al.2026 · 0 citationsarXiv
Design-Based Anytime-Valid Inference for Randomized Experiments with Delayed Outcomes and Staggered Entry2603.25971v2 · Michael Lindon, Nathan Kallus2026 · 0 citationsarXiv
Efficient Inference after Directionally Stable Adaptive Experiments2602.21478v1 · Zikai Shen, Houssam Zenati, Nathan Kallus et al.2026 · 0 citationsarXiv
GAAVI: Global Asymptotic Anytime Valid Inference for the Conditional Mean Function2602.08096v1 · Brian M Cho, Raaz Dwivedi, Nathan Kallus2026 · 0 citationsarXiv
Causal Inference on Networks under Misspecified Exposure Mappings: A Partial Identification Framework2602.03459v1 · Maresa Schroder, Miruna Oprescu, Stefan Feuerriegel et al.2026 · 0 citationsarXiv
Exploration in the Limit2601.00084v1 · Brian M. Cho, Nathan Kallus2025 · 0 citationsarXiv
Soft Fitted Q-Iteration without Bellman Completeness: Occupancy Reweighting and Temperature Annealing2512.23927v3 · Lars van der Laan, Nathan Kallus2025 · 0 citationsarXiv
Efficient Inference for Inverse Reinforcement Learning and Dynamic Discrete Choice Models2512.24407v2 · Lars van der Laan, Aur'elien Bibaut, Nathan Kallus2025 · 0 citationsarXiv
Bellman Calibration for -Learning in Offline Reinforcement Learning2512.23694v2 · Lars van der Laan, Nathan Kallus2025 · 0 citationsarXiv
Fitted Q-Evaluation without Bellman Completeness via Occupancy Weighting2512.23805v4 · Lars van der Laan, Nathan Kallus2025 · 0 citationsarXiv
Semiparametric Preference Optimization: Your Language Model is Secretly a Single-Index Model2512.21917v3 · Nathan Kallus2025 · 0 citationsarXiv
Experimentation Under Non-stationary Interference2511.06685v1 · Su Jia, Peter Frazier, Nathan Kallus et al.2025 · 0 citationsarXiv
The Value of Personalized Recommendations: Evidence from Netflix2511.07280v5 · Kevin Zielnicki, Guy Aridor, Aur'elien Bibaut et al.2025 · 2 citationsarXiv
Rank-GRPO: Training LLM-based Conversational Recommender Systems with Reinforcement Learning2510.20150v5 · Yaochen Zhu, Harald Steck, Dawen Liang et al.2025 · 0 citationsarXiv
Fast Best-in-Class Regret for Contextual Bandits2510.15483v3 · Samuel Girard, Aurelien Bibaut, Arthur Gretton et al.2025 · 0 citationsarXiv
Does Weighting Improve Matrix Factorization for Recommender Systems?2510.10440v1 · Alex Ayoub, Samuel Robertson, Dawen Liang et al.2025 · 0 citationsarXiv
DiFFPO: Training Diffusion LLMs to Reason Fast and Furious via Reinforcement Learning2510.02212v2 · Hanyang Zhao, Dawen Liang, Wenpin Tang et al.2025 · 0 citationsarXiv
Entropy After </Think> for reasoning model early exiting2509.26522v3 · Xi Wang, James McInerney, Lequn Wang et al.2025 · 0 citationsarXiv
Inverse Reinforcement Learning with Just Classification and a Few Regressions2509.21172v2 · Lars van der Laan, Nathan Kallus, Aurelien Bibaut2025 · 0 citationsarXiv
Simulation-Based Inference for Adaptive Experiments2506.02881v1 · Brian M Cho, Aur'elien Bibaut, Nathan Kallus2025 · 0 citationsarXiv
Optimization of Epsilon-Greedy Exploration2506.03324v1 · Ethan Che, Hakan Ceylan, James McInerney et al.2025 · 0 citationsarXiv
Value-Guided Search for Efficient Chain-of-Thought Reasoning2505.17373v2 · Kaiwen Wang, Jin Peng Zhou, Jonathan Chang et al.2025 · 0 citationsarXiv
Efficient Adaptive Experimentation with Noncompliance2505.17468v2 · Miruna Oprescu, Brian M Cho, Nathan Kallus2025 · 0 citationsarXiv
Nonparametric Instrumental Variable Inference with Many Weak Instruments2505.07729v3 · Lars van der Laan, Nathan Kallus, Aur'elien Bibaut2025 · 0 citationsarXiv
An Empirical Study on Zero-Data Bootstrapping for Conversational Recommender Systems2504.15476v2 · Rohan Surana, Junda Wu, Zhouhang Xie et al.2025 · 1 citationarXiv
SNPL: Simultaneous Policy Learning and Evaluation for Safe Multi-Objective Policy Improvement2503.12760v2 · Brian Cho, Ana-Roxana Pop, Ariel Evnine et al.2025 · 0 citationsarXiv
: Provably Optimal Distributional RL for LLM Post-Training2502.20548v2 · Jin Peng Zhou, Kaiwen Wang, Jonathan Chang et al.2025 · 0 citationsarXiv
Collaborative Retrieval for Large Language Model-based Conversational Recommender Systems2502.14137v1 · Yaochen Zhu, Chao Wan, Harald Steck et al.2025 · 10 citationsarXiv
Evaluating Decision Rules Across Many Weak Experiments2502.08763v2 · Winston Chou, Colin Gray, Nathan Kallus et al.2025 · 1 citationarXiv
GST-UNet: A Neural Framework for Spatiotemporal Causal Inference with Time-Varying Confounding2502.05295v2 · Miruna Oprescu, David K. Park, Xihaier Luo et al.2025 · 0 citationsarXiv
Automatic Debiased Machine Learning for Smooth Functionals of Nonparametric M-Estimands2501.11868v3 · Lars van der Laan, Aurelien Bibaut, Nathan Kallus et al.2025 · 0 citationsarXiv
Semiparametric Double Reinforcement Learning with Applications to Long-Term Causal Inference2501.06926v5 · Lars van der Laan, David Hubbard, Allen Tran et al.2025 · 0 citationsarXiv
Reward Maximization for Pure Exploration: Minimax Optimal Good Arm Identification for Nonparametric Multi-Armed Bandits2410.15564v1 · Brian Cho, Dominik Meier, Kyra Gan et al.2024 · 0 citationsarXiv
Anytime-Valid Continuous-Time Confidence Processes for Inhomogeneous Poisson Processes2410.09282v1 · Michael Lindon, Nathan Kallus2024 · 0 citationsarXiv
Career total: 222 works. 172 are in this corpus.Showing the 50 most recent.
Profile built from the corpus for this byline.
Author records are still filling in while the Hub is in alpha. If this is your page, you'll be able to claim it soon. Spot a mistake? Tell us.