Pith. sign in

REVIEW 4 major objections 4 minor 1 cited by

TAB-Fields: A Maximum Entropy Framework for Mission-Aware Adversarial Planning

T0 review · 4 major / 4 minor · reviewed 2026-08-11 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read TAB-Fields replace unknown adversary policies with a maximum-entropy distribution over mission-consistent trajectories, and the resulting planner outperforms policy-assuming baselines.

desk verdict The max-entropy core is textbook, but the paper's real novelty—using marginals as a surrogate transition model in POMCP—rests on a belief update that is not Bayesian and likely explains much of the reported gains. read the letter →

arxiv 2412.02570 v1 pith:RXUWIENB submitted 2024-12-03 cs.RO cs.AIcs.LGcs.MAcs.SYeess.SY

Gokul Puthumanaillam , Jae Hyuk Song , Nurzhan Yesmagambet , Shinkyu Park , Melkior Ornik This is my paper · ORCID
classification cs.ROcs.AIcs.LGcs.MAcs.SYeess.SY
keywords AdversarialplanningMission-constrainedunderuncertaintyMaximumentropyPOMCPBelief-spaceUnknowntransitiondynamicsInterception
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper argues that an autonomous agent can plan against an adversary without knowing the adversary's policy, as long as it knows the adversary's mission and environment. It builds Task-Aware Behavior Fields (TAB-Fields), time-indexed distributions over adversary states obtained by minimizing KL divergence from a physically feasible reference process subject to mission constraints. These marginal distributions are plugged into POMCP as a substitute for the unknown transition dynamics, yielding TAB-conditioned POMCP. In simulation and hardware, the resulting agent intercepts the adversary faster and more often than baselines that assume a fixed policy, learn from data, or ignore mission constraints. If the effect is robust, mission specifications alone can carry adversarial reasoning in domains where behavior data is unavailable.

What carries the argument

The central object is the Task-Aware Behavior Field (TAB-Field), a collection of per-timestep marginal distributions over adversary states. It is the solution of $\min_{P} D_{KL}(P \| Q)$ subject to an initial-state constraint, mission constraints of the form $\mathbb{E}_{P}[f_{M}(s^{adv}_{0:T})] = c_{M}$, and zero probability on forbidden states; its optimal form is $P^{*}(s^{adv}_{0:T}) = \frac{1}{Z} Q(s^{adv}_{0:T}) \exp(-\lambda^{T} f_{M}(s^{adv}_{0:T}))$. Here $Q$ is a reference process assigning probability only to physically feasible paths. The marginals are computed by iterative inference in probabilistic graphical models. The second piece is TAB-conditioned POMCP, which uses those marginals to propagate particles and update beliefs in place of the unknown adversary transition model.

What would settle it

Run an occlusion experiment with an adversary following a known, fixed policy: record where it actually emerges after a long unobserved stretch and compare this empirical distribution with the TAB-Field marginal; if the marginal assigns low probability to states the adversary reaches often, the surrogate replacement is false.

Watch

Extended reading notes

Core claim

The central claim is that the minimum-KL (maximum entropy) distribution over adversary trajectories, constrained only by mission specifications and physical feasibility, is a sufficient stand-in for the adversary's unknown transition dynamics in belief-space planning. Concretely, the paper replaces the Bayesian convolution in the belief update with the TAB-Field marginal $P(s^{adv}_{t+1})$ whenever no checkpoint observation arrives, and samples next adversary states from that marginal inside POMCP. The reported consequence is consistently lower Adversary Task Completion Rate and fewer steps to interception than uniform-motion, fixed-shortest-path, and MLE-learned baselines across five mission types, with the advantage growing as missions become more complex. The paper frames this as avoiding both the brittleness of specific policy assumptions and the data hunger of learned behavior models.

Load-bearing premise

The load-bearing premise is that the TAB-Field marginal, computed without any adversary policy or behavior data, faithfully replaces the adversary's true unknown transition dynamics in the belief update whenever no fresh observation arrives.

Editorial extensions

If this is right

  • An ego agent can plan interception, avoidance, or surveillance missions with no adversary policy model and no behavioral training data, using only the adversary's mission and environment.
  • TAB-conditioned POMCP reduces average steps to interception and adversary task completion relative to uniform-motion, fixed-policy, and MLE baselines across the five reported mission types, in both ground-robot hardware and underwater simulation.
  • The performance advantage over baselines widens as mission complexity grows, indicating that mission constraints become more informative as the feasible behavior space expands.
  • The representation is planner-agnostic: any POMDP solver that accepts a time-indexed transition distribution could in principle be conditioned on TAB-Fields, with only about a 1.4x computational overhead demonstrated for POMCP.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • An extension left implicit is that TAB-Fields could act as a reusable prior in receding-horizon planners: whenever the adversary's mission tuple changes, only the entropy optimization needs to be recomputed, not the planner's policy.
  • The same field could be consumed by trajectory predictors or risk maps, not only planners, since it is a standalone time-indexed distribution over adversary states.
  • The belief update's substitution of the TAB-Field marginal for the Bayesian convolution when no observation arrives is an approximation the paper does not flag; its reliability under long occlusions or with a deliberately misleading adversary policy is a natural stress test.
  • A quantitative bound linking constraint tightness to the worst-case gap between the maximum-entropy distribution and the true adversary distribution would turn the method's empirical advantage into a more general guarantee.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 4 minor

Summary. The paper introduces TAB-Fields, a representation of an adversary's possible state over time, computed as the minimum-KL (maximum entropy) distribution over adversary trajectories subject to known mission and environmental constraints. The authors propose TAB-conditioned POMCP, which adapts POMCP by replacing the adversary's unknown transition dynamics with the time-marginals of the TAB-Field distribution, both in the belief update and in particle propagation. The paper claims superior performance over baselines that assume specific adversary policies or ignore mission constraints, based on ground-robot hardware experiments and underwater vehicle simulations. The manuscript also provides a project website with code and videos.

Significance. If the central claim were sound, the approach would be valuable: it would allow an ego agent to plan against an adversary using only mission specifications and environment constraints, without adversary policy models or behavior data. The optimization problem in Eq. (1) is clearly stated and the exponential-family solution is standard. The paper ships code and evaluation videos, which is a strength. However, the significance is currently undermined because the integration of TAB-Fields into POMCP uses a non-Bayesian belief update and state sampling that disregards the current belief and particle state. The reported performance gains may be an artifact of this prior-resetting heuristic rather than evidence that TAB-Fields capture adversary behavior. The lack of statistical detail in the experiments further weakens the empirical claim.

major comments (4)
  1. [Section 4.1, belief update equation] The belief update b_{t+1}(s^adv) = η O(o^adv|s^adv) P(s^adv) is not a Bayesian posterior. A correct prediction step would convolve the current belief b_t with a transition kernel, e.g., the conditional distribution derived from the same min-KL path measure. Replacing it with the static time-marginal P(s^adv) discards the information in b_t and the last checkpoint observation, and resets the belief to the mission-constrained prior at every step. The paper does not flag this as an approximation. This is load-bearing because the abstract and Section 4 claim that TAB-Fields serve as a principled substitute for the unknown transition dynamics. The authors should either derive the update from the path measure's conditional transitions or explicitly characterize this as a heuristic and provide evidence that the heuristic does not drive the reported gains.
  2. [Section 4.1, TAB-POMCP particle propagation] In TAB-POMCP, "the next adversary state is sampled from the TAB-Field distribution" independently of the current particle state. This allows particles to teleport to mission-consistent regions that are physically unreachable from their current state, so the tree search is not a valid belief-space POMCP over the original state space. The resulting value estimates correspond to a different process in which the adversary's state is independent across time. Since the reported performance advantage could come from this prior concentration rather than from the max-entropy model, the central empirical claim is not yet supported.
  3. [Section 5.3, Tables 1 and 2] Tables 1 and 2 report ATCR and StI values without error bars, standard deviations, or significance tests, despite stating that results are averaged over 150 experiments per mission type. The claim that TAB-POMCP "consistently outperforms all baselines" is not statistically supported. Additionally, the baselines are under-specified: MLE-POMCP's data source, estimation procedure, and hyperparameters are not described, and FP-POMCP's exact shortest-path implementation is not given. Without this information, it is difficult to assess whether the comparison is fair or whether the baselines are deliberately weak.
  4. [Section 4, marginal computation paragraph] The paper states that the marginals P*(s^adv_t) "can be efficiently computed" using iterative algorithms from probabilistic graphical models, but provides no algorithm, no convergence guarantees, and no complexity analysis. The scalability claim in Section 5.3, including the implicit dimensionality reduction, rests on this assertion, yet the only quantitative evidence is a single runtime ratio (1.4x) with no details on state-space size, discretization, or computational environment. This is a missing support for a central practical claim.
minor comments (4)
  1. [Figure 3 caption] The caption contains the typo "TAB-POCMP" in three places; it should read "TAB-POMCP".
  2. [Section 5.2, paragraph after Figure 3] The sentence "TAB-POMCP does not assume specific behavior patterns anded by known constraints" contains an obvious typo; it should likely read "and is guided by known constraints."
  3. [Section 5.3, M1-M5 mission descriptions] The acronym ATCR is used with a downward arrow but never defined; the text should explain that lower ATCR (adversary task completion rate) indicates better performance for the ego agent, as the metric measures the adversary's success rate.
  4. [References] The citation of Jie et al. (2017) as prior work on converting natural language mission specifications to constraint tuples is questionable, as that paper addresses named entity recognition. If a more appropriate reference exists, it should be cited; otherwise the sentence overstates the support.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the max-entropy TAB-Fields are derived from external mission/environment constraints, and the few self-citations are not load-bearing.

full rationale

The derivation chain is self-contained. Equation (1) defines the distribution P* by minimizing D_KL(P||Q) subject to initial-state, mission-constraint, and environment-constraint conditions; the resulting exponential-family solution is credited to textbook results (Thomas and Joy, 2006) and to the externally established maximum-entropy principle (Jaynes, 1957, 1982), not to a self-citation. TAB-Fields are then the time-marginals P*(s_t) computed with standard graphical-model algorithms, and they are inserted into the POMCP belief update in Section 4.1. No parameter of the field is fitted to the ATCR/StI outcomes, and the mission constraints f_M are inputs rather than quantities learned from the benchmark results. The Section 4.1 update b_{t+1}=eta O(s')P(s') and the TAB-POMCP step of sampling 'the next adversary state from the TAB-Field distribution' are approximations that replace the Bayesian convolution with a surrogate, and they do discard b_t when no observation arrives; this is a modeling and correctness concern, not a circular equivalence, because the prediction is not defined in terms of the belief or the reward it is used to optimize. The only author self-citations (Puthumanaillam et al. 2024a,b,c) concern natural-language-to-constraint conversion, time-varying POMDP learning, and a navigation package in the hardware setup; none carries the central max-entropy derivation. The empirical setup gives the ego agent the adversary's true mission constraints, which is an evaluation-validity point rather than a reduction of the derivation to its own inputs.

Assumptions & free parameters 0 free parameters · 6 assumptions · 0 invented entities

The method introduces no fitted numeric parameters and no invented physical entities. It rests on domain assumptions about mission specification, the reference process Q, and the validity of replacing the transitioned belief with the TAB-Field marginal; the last is an unstated approximation that should be exposed and tested.

assumptions (6)
  • domain assumption Known mission constraints fully specify the relevant information about adversary behavior; no other biases should be introduced.
    Section 4 builds P only from the initial distribution, mission constraints, and obstacle avoidance; the entire max entropy argument assumes these constraints are the only relevant facts.
  • domain assumption The reference distribution Q over physically feasible trajectories is a reasonable prior for adversary motion.
    Eq. (1) minimizes D_KL(P||Q); if Q is far from the adversary's actual motion class, the tilted distribution inherits that error.
  • domain assumption Natural language missions can be converted losslessly into constraint tuples M and each tuple yields a valid indicator constraint f_M.
    Section 3 delegates conversion to prior work and assumes the conversion preserves all mission-relevant constraints.
  • ad hoc to paper The TAB-Field marginal can stand in for the transitioned belief in the POMDP belief update.
    Section 4.1 uses b_{t+1} = eta O(...) P(...) instead of the standard Bayesian convolution; this is presented without justification as an approximation.
  • domain assumption The environment is static; obstacles and checkpoints do not move.
    The limitations section states the formulation does not yet account for dynamic obstacles, so the paper's analysis is conditional on a static world.
  • standard math The solution to the constrained KL minimization takes the exponential family form and marginals can be computed with iterative graphical model algorithms.
    Eq. (2) relies on the classical exponential family result; this is standard but used without proof, and the algorithmic claim is not backed by details.

how reviews work

0 comments
Cite this review

Pith. "Pith review of TAB-Fields: A Maximum Entropy Framework for Mission-Aware Adversarial Planning." pith.science (2026). https://pith.science/paper/RXUWIENB

@misc{pith2026241202570,
  author       = {Pith},
  title        = {Pith review of: TAB-Fields: A Maximum Entropy Framework for Mission-Aware Adversarial Planning},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/RXUWIENB}},
  note         = {Machine review of arXiv:2412.02570}
}
read the original abstract

Autonomous agents operating in adversarial scenarios face a fundamental challenge: while they may know their adversaries' high-level objectives, such as reaching specific destinations within time constraints, the exact policies these adversaries will employ remain unknown. Traditional approaches address this challenge by treating the adversary's state as a partially observable element, leading to a formulation as a Partially Observable Markov Decision Process (POMDP). However, the induced belief-space dynamics in a POMDP require knowledge of the system's transition dynamics, which, in this case, depend on the adversary's unknown policy. Our key observation is that while an adversary's exact policy is unknown, their behavior is necessarily constrained by their mission objectives and the physical environment, allowing us to characterize the space of possible behaviors without assuming specific policies. In this paper, we develop Task-Aware Behavior Fields (TAB-Fields), a representation that captures adversary state distributions over time by computing the most unbiased probability distribution consistent with known constraints. We construct TAB-Fields by solving a constrained optimization problem that minimizes additional assumptions about adversary behavior beyond mission and environmental requirements. We integrate TAB-Fields with standard planning algorithms by introducing TAB-conditioned POMCP, an adaptation of Partially Observable Monte Carlo Planning. Through experiments in simulation with underwater robots and hardware implementations with ground robots, we demonstrate that our approach achieves superior performance compared to baselines that either assume specific adversary policies or neglect mission constraints altogether. Evaluation videos and code are available at https://tab-fields.github.io.

Figures

Figures reproduced from arXiv: 2412.02570 by the authors.

Figure 1
Figure 1. Overview of the proposed approach applied to an interception task. The adversary’s task is [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Example mission and its TAB-Field, where darker areas indicate higher probability of adversary pres￾ence. Red area denotes adversary start position and purple area indicates the goal checkpoint. The optimization in equation (1) extends techniques from maximum entropy IRL (Ziebart et al., 2008), where sim￾ilar techniques are used to model expert behavior with￾out assuming specific reward functions. In equation (1), D… view at source ↗
Figure 3
Figure 3. Comparison of agent (green) and adversary (red) trajectories followed by different ap [PITH_FULL_IMAGE:figures/full_fig_p009_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: Agent (green) and adversary (red) tra￾jectories using TAB-POMCP. Teal bubbles indi￾cate checkpoints. Adversary task: Reach corals after visiting checkpoints 1, 2, 3 in order, taking no more than 30s between checkpoints. As shown in [PITH_FULL_IMAGE:figures/full_fig_p0…

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Enhancing Robot Navigation Policies with Task-Specific Uncertainty Managements

    cs.RO 2025-05 conditional novelty 5.0 of 10

    GUIDE conditions a soft actor-critic navigation policy on task-specific uncertainty maps built from language-specified tasks and reports large gains over baselines in lake experiments.

Reference graph

Works this paper leans on

52 extracted references · 47 canonical work pages · cited by 1 Pith paper

  1. [1]

    Exploration and apprenticeship learning in reinforcement learning

    Pieter Abbeel and Andrew Y Ng. Exploration and apprenticeship learning in reinforcement learning. In International Conference on Machine Learning, 2005

    work page 2005

  2. [2]

    Adversarial deep reinforcement learning to mitigate sensor and communication attacks for secure swarm robotics

    Mahmoud Abouelyazid. Adversarial deep reinforcement learning to mitigate sensor and communication attacks for secure swarm robotics. Journal of Intelligent Connectivity and Emerging Technologies, 2023

    work page 2023

  3. [3]

    A survey of inverse reinforcement learning

    Stephen Adams, Tyler Cody, and Peter A Beling. A survey of inverse reinforcement learning. Artificial Intelligence Review, pages 4307--4346, 2022

    work page 2022

  4. [4]

    Maximum entropy inverse reinforcement learning in continuous state spaces with path integrals

    Navid Aghasadeghi and Timothy Bretl. Maximum entropy inverse reinforcement learning in continuous state spaces with path integrals. In IEEE/RSJ International Conference on Intelligent Robots and Systems, 2011

    work page 2011

  5. [5]

    Robotic strategic behavior in adversarial environments

    Noa Agmon. Robotic strategic behavior in adversarial environments. In International Joint Conference on Artificial Intelligence, 2017

    work page 2017

  6. [6]

    A survey of inverse reinforcement learning: Challenges, methods and progress

    Saurabh Arora and Prashant Doshi. A survey of inverse reinforcement learning: Challenges, methods and progress. Artificial Intelligence, 2021

    2021

  7. [7]

    Near-optimal regret bounds for reinforcement learning

    Peter Auer, Thomas Jaksch, and Ronald Ortner. Near-optimal regret bounds for reinforcement learning. In Neural Information Processing Systems, 2008

    work page 2008

  8. [8]

    Partially observable Monte Carlo planning with state variable constraints for mobile robot navigation

    Alberto Castellini, Enrico Marchesini, and Alessandro Farinelli. Partially observable Monte Carlo planning with state variable constraints for mobile robot navigation. Engineering Applications of Artificial Intelligence, 2021

    work page 2021

Show all 52 references
  1. [9]

    On the complexity of computing maximum entropy for Markovian models

    Taolue Chen and Tingting Han. On the complexity of computing maximum entropy for Markovian models. Leibniz International Proceedings in Informatics, 2014

    2014

  2. [10]

    Exploration-exploitation trade-off in reinforcement learning on online markov decision processes with global concave rewards

    Wang Chi Cheung. Exploration-exploitation trade-off in reinforcement learning on online markov decision processes with global concave rewards. arXiv preprint arXiv:1905.06466, 2019

    1905 arXiv

  3. [11]

    Decentralized mcts via learned teammate models

    Aleksander Czechowski and Frans A Oliehoek. Decentralized mcts via learned teammate models. In International Conference on International Joint Conferences on Artificial Intelligence, 2021

    2021

  4. [12]

    Target surveillance in adversarial environments using pomdps

    Maxim Egorov, Mykel Kochenderfer, and Jaak Uudmae. Target surveillance in adversarial environments using pomdps. In AAAI Conference on Artificial Intelligence, 2016

    2016

  5. [13]

    A comprehensive survey on safe reinforcement learning

    Javier Garc a and Fernando Fern \'a ndez. A comprehensive survey on safe reinforcement learning. Journal of Machine Learning Research, 16 0 (1): 0 1437--1480, 2015

    2015

  6. [14]

    Multi-agent deep reinforcement learning: A survey

    S Gronauer and K Diepold. Multi-agent deep reinforcement learning: A survey. Artificial Intelligence Review, 2022

    2022

  7. [15]

    Towards modeling the behavior of autonomous systems and humans for trusted operations

    Weiqing Gu, Ranjeev Mittu, Julie Marble, Gavin Taylor, Ciara Sibley, Joseph Coyne, and William F Lawless. Towards modeling the behavior of autonomous systems and humans for trusted operations. In AAAI Spring Symposium Series, 2014

    2014

  8. [16]

    Learning others' intentional models in multi-agent settings using interactive POMDP s

    Yanlin Han and Piotr Gmytrasiewicz. Learning others' intentional models in multi-agent settings using interactive POMDP s. In Neural Information Processing Systems, 2018

    2018

  9. [17]

    I POMDP -net: A deep neural network for partially observable multi-agent planning using interactive POMDP s

    Yanlin Han and Piotr Gmytrasiewicz. I POMDP -net: A deep neural network for partially observable multi-agent planning using interactive POMDP s. In AAAI Conference on Artificial Intelligence, 2019

    2019

  10. [18]

    A survey of multi-robot regular and adversarial patrolling

    Li Huang, MengChu Zhou, Kuangrong Hao, and Edwin Hou. A survey of multi-robot regular and adversarial patrolling. IEEE/CAA Journal of Automatica Sinica, 2019

    2019

  11. [19]

    Robust dynamic programming

    Garud N Iyengar. Robust dynamic programming. Mathematics of Operations Research, 2005

    2005

  12. [20]

    Information theory and statistical mechanics

    Edwin T Jaynes. Information theory and statistical mechanics. Physical Review, 1957

    1957

  13. [21]

    On the rationale of maximum-entropy methods

    Edwin T Jaynes. On the rationale of maximum-entropy methods. Proceedings of the IEEE, 1982

    1982

  14. [22]

    Efficient dependency-guided named entity recognition

    Zhanming Jie, Aldrian Muis, and Wei Lu. Efficient dependency-guided named entity recognition. In AAAI Conference on Artificial Intelligence, 2017

    2017

  15. [23]

    Planning and acting in partially observable stochastic domains

    Leslie Pack Kaelbling, Michael L Littman, and Anthony R Cassandra. Planning and acting in partially observable stochastic domains. Artificial Intelligence, 1998

    1998

  16. [24]

    Probabilistic Graphical Models: Principles and Techniques , 2009

    Daphane Koller. Probabilistic Graphical Models: Principles and Techniques , 2009

    2009

  17. [25]

    Review of pedestrian trajectory prediction methods: Comparing deep learning and knowledge-based approaches

    Raphael Korbmacher and Antoine Tordeux. Review of pedestrian trajectory prediction methods: Comparing deep learning and knowledge-based approaches. IEEE Transactions on Intelligent Transportation Systems, 2022

    2022

  18. [26]

    Partially observable markov decision processes in robotics: A survey

    Mikko Lauri, David Hsu, and Joni Pajarinen. Partially observable markov decision processes in robotics: A survey. IEEE Transactions on Robotics, 2022

    2022

  19. [27]

    A survey of the S chr \"o dinger problem and some of its connections with optimal transport

    Christian L \'e onard. A survey of the S chr \"o dinger problem and some of its connections with optimal transport. arXiv preprint arXiv:1308.0215, 2013

    2013 arXiv

  20. [28]

    o dinger

    Simone Di Marino and Augusto Gerolin. An optimal transport approach for the S chr \"o dinger" odinger bridge problem and convergence of S inkhorn algorithm. Journal of Scientific Computing, 2020

    2020

  21. [29]

    Towards applying interactive pomdps to real-world adversary modeling

    Brenda Ng, Carol Meyers, Kofi Boakye, and John Nitao. Towards applying interactive pomdps to real-world adversary modeling. In AAAI Conference on Artificial Intelligence, 2010

    2010

  22. [30]

    Robust control of M arkov decision processes with uncertain transition matrices

    Arnab Nilim and Laurent El Ghaoui. Robust control of M arkov decision processes with uncertain transition matrices. Operations Research, 2005

    2005

  23. [31]

    Reasoning in Uncertain Adversarial Environments in Agent/Multiagent Systems

    Praveen Paruchuri. Reasoning in Uncertain Adversarial Environments in Agent/Multiagent Systems. PhD thesis, Ph.D. Dissertation Proposal, 2007

    2007

  24. [32]

    Predicting actions to act predictably: Cooperative partial motion planning with maximum entropy models

    Mark Pfeiffer, Ulrich Schwesinger, Hannes Sommer, Enric Galceran, and Roland Siegwart. Predicting actions to act predictably: Cooperative partial motion planning with maximum entropy models. In IEEE/RSJ International Conference on Intelligent Robots and Systems, 2016

    2016

  25. [33]

    Weathering ongoing uncertainty: Learning and planning in a time-varying partially observable environment

    Gokul Puthumanaillam, Xiangyu Liu, Negar Mehr, and Melkior Ornik. Weathering ongoing uncertainty: Learning and planning in a time-varying partially observable environment. In IEEE International Conference on Robotics and Automation, 2024 a

    2024

  26. [34]

    Enhancing robot navigation policies with task-specific uncertainty management

    Gokul Puthumanaillam, Paulo Padrao, Jose Fuentes, Leonardo Bobadilla, and Melkior Ornik. Enhancing robot navigation policies with task-specific uncertainty management. arXiv preprint arXiv:2410.15178, 2024 b

    2024

  27. [35]

    ComTraQ-MPC: M eta-trained DQN-MPC integration for trajectory tracking with limited active localization updates

    Gokul Puthumanaillam, Manav Vora, and Melkior Ornik. ComTraQ-MPC: M eta-trained DQN-MPC integration for trajectory tracking with limited active localization updates. arXiv preprint arXiv:2403.01564, 2024 c

    2024 arXiv

  28. [36]

    Adversarial models for opponent intent inferencing

    Eugene Santos Jr and Qunhua Zhao. Adversarial models for opponent intent inferencing. Adversarial Reasoning: C omputational Approaches to Reading the Opponents Mind , 2006

    2006

  29. [37]

    Modeling adversarial intent for interactive simulation and gaming: the fused intent system

    Eugene Santos Jr, Bruce McQueary, and Lee Krause. Modeling adversarial intent for interactive simulation and gaming: the fused intent system. In Modeling and Simulation for Military Operations, 2008

    2008

  30. [38]

    Entropy maximization for constrained Markov decision processes

    Yagiz Savas, Melkior Ornik, Murat Cubuktepe, and Ufuk Topcu. Entropy maximization for constrained Markov decision processes. In Allerton Conference on Communication, Control, and Computing, 2018

    2018

  31. [39]

    Entropy maximization for Markov decision processes under temporal logic constraints

    Yagiz Savas, Melkior Ornik, Murat Cubuktepe, Mustafa O Karabag, and Ufuk Topcu. Entropy maximization for Markov decision processes under temporal logic constraints. IEEE Transactions on Automatic Control, 2019

    2019

  32. [40]

    A survey of point-based POMDP solvers

    Guy Shani, Joelle Pineau, and Robert Kaplow. A survey of point-based POMDP solvers. Autonomous Agents and Multi-Agent Systems, 2013

    2013

  33. [41]

    Monte-carlo planning in large POMDP s

    David Silver and Joel Veness. Monte-carlo planning in large POMDP s. In Neural Information Processing Systems, 2010

    2010

  34. [42]

    Despot: O nline POMDP planning with regularization

    Adhiraj Somani, Nan Ye, David Hsu, and Wee Sun Lee. Despot: O nline POMDP planning with regularization. In Neural Information Processing Systems, 2013

    2013

  35. [43]

    Elements of Information Theory

    MTCAJ Thomas and A Thomas Joy. Elements of Information Theory. Wiley-Interscience, 2006

    2006

  36. [44]

    Monte carlo pomdps

    Sebastian Thrun. Monte carlo pomdps. In Neural Information Processing Systems, 1999

    1999

  37. [45]

    Efficient computation of optimal actions

    Emanuel Todorov. Efficient computation of optimal actions. National Academy of Sciences, 2009

    2009

  38. [46]

    Robust markov decision processes

    Wolfram Wiesemann, Daniel Kuhn, and Ber c Rustem. Robust markov decision processes. Mathematics of Operations Research, 2013

    2013

  39. [47]

    Robust M arkov D ecision P rocesses without model estimation

    Wenhao Yang, Han Wang, Tadashi Kozuno, Scott M Jordan, and Zhihua Zhang. Robust M arkov D ecision P rocesses without model estimation. arXiv preprint arXiv:2302.01248, 2023

    2023 arXiv

  40. [48]

    Robust deep reinforcement learning against adversarial perturbations on state observations

    Huan Zhang, Hongge Chen, Chaowei Xiao, Bo Li, Mingyan Liu, Duane Boning, and Cho-Jui Hsieh. Robust deep reinforcement learning against adversarial perturbations on state observations. In Neural Information Processing Systems, 2020

    2020

  41. [49]

    Multi-robot coordination and planning in uncertain and adversarial environments

    Lifeng Zhou and Pratap Tokekar. Multi-robot coordination and planning in uncertain and adversarial environments. Current Robotics Reports, 2021

    2021

  42. [50]

    Maximum entropy inverse reinforcement learning

    Brian D Ziebart, Andrew L Maas, J Andrew Bagnell, Anind K Dey, et al. Maximum entropy inverse reinforcement learning. In AAAI Conference on Artificial Intelligence, 2008

    2008

  43. [51]

    Planning-based prediction for pedestrians

    Brian D Ziebart, Nathan Ratliff, Garratt Gallagher, Christoph Mertz, Kevin Peterson, J Andrew Bagnell, Martial Hebert, Anind K Dey, and Siddhartha Srinivasa. Planning-based prediction for pedestrians. In IEEE/RSJ International Conference on Intelligent Robots and Systems, 2009

    2009

  44. [52]

    The adversarial activity model for bounded rational agents

    Inon Zuckerman, Sarit Kraus, and Jeffrey S Rosenschein. The adversarial activity model for bounded rational agents. Autonomous Agents and Multi-Agent Systems, 2012

    2012

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.