Pith. sign in

REVIEW 3 major objections 3 minor 34 references

A Forward-Inverse Dynamic Game Framework for Enhanced Multi-Agent Trajectory Planning

T0 review · 3 major / 3 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read This paper claims that a state-dependent KL-regularized dynamic game, solved by freezing the regularization coefficient along a nominal trajectory, has a unique Gaussian feedback Nash equilibrium at each iteration and that the adaptive…

desk verdict The state-dependent KL weighting is a genuinely good idea, but the central FBNE theorem has a load-bearing sign error and the paper's own collision cost violates the theorem's PSD premise; this version should be rejected, though the idea deserves a serious revised look. read the letter →

arxiv 2608.01636 v1 pith:FFZKZTZA submitted 2026-08-03 cs.RO

Tianle Liu , Youcheng Niu , Jing Zeng , Shuo Li , Jinming Xu This is my paper · ORCID
classification cs.RO
keywords multi-agenttrajectoryplanningfeedbackNashequilibriumKL-regularizeddynamicgamestate-dependentregularizationmaximum-entropyinversereinforcementlearningscene-awarecostinferencecollisionavoidance
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

This paper tries to establish that multi-agent trajectory planning under unknown goals can be treated as a forward–inverse dynamic game in which a state-dependent KL-regularization weight balances optimal control against a learned reference policy. The forward game is solved by linearizing dynamics and costs around a nominal trajectory and freezing the regularization coefficient per iteration, yielding a local linear–quadratic game. The paper claims each such local game has a unique Gaussian feedback Nash equilibrium with well-defined covariances and value recursions (Proposition 1), and that the adaptive weight is Lipschitz in the joint state (Proposition 2). The inverse module infers each agent's cost weights from demonstrations using maximum-entropy IRL with a scene-aware network and a physics-informed rule loss. If correct, this gives a principled way to make game-theoretic planners safer in dense, safety-critical scenes without committing to full rationality or fixed imitation strength.

What carries the argument

The mechanism that carries the argument is the state-dependent regularization coefficient $\lambda_i(x_t) = \lambda_i^{\min} + (\lambda_i^{\max} - \lambda_i^{\min}) \exp\!\left(-(d_i^{\mathrm{obs}}(x_t))^2/(2\sigma^2)\right)$, where $d_i^{\mathrm{obs}}(x_t)$ is the distance from agent $i$ to the nearest obstacle or other agent. Evaluated on the nominal trajectory and frozen during each inner Riccati recursion, this coefficient makes the local game quadratic and well-posed while letting proximity tighten or relax adherence to the reference policy. The analytic Gaussian policy from Lemma 1 and the coupled Riccati-style recursions for the gains and value parameters are the objects that carry the equilibrium computation.

What would settle it

Evaluate the Hessian of the collision-avoidance cost $f_2 = -\log(\lVert p_i - p_j \rVert^2)$ at a typical near-collision configuration; if it has a negative eigenvalue, the Proposition 1 premise $Q_{i,t} \succeq 0$ fails. Then run the two-agent value recursion (13a) with that cost: if $Z_{i,t}$ loses positive semidefiniteness at any step, the claimed unique Gaussian FBNE per iteration is not assured for the planner as implemented.

Watch

Extended reading notes

Core claim

The central claim is that adding a state-dependent weight $\lambda_i(x_t)$ to the KL-regularized dynamic game preserves the tractable structure needed for feedback Nash equilibrium computation while adapting imitation strength to local risk. The paper derives a Gaussian feedback equilibrium for the frozen local game: each agent's optimal policy is Gaussian with mean $-P_{i,t}\delta x_t - \alpha_{i,t}$ and covariance $\lambda_{i,t}(M_{i,k,t})^{-1}$, and it proves per-iteration well-posedness and Lipschitz continuity of the weight. Alongside this, the inverse module recovers cost weights from demonstrated trajectories via a scene-aware network trained with maximum-entropy IRL and a rule penalty, and the experiments report lower collision counts and trajectory errors than fixed-$\lambda$ and KKT-based baselines, including zero collisions in a five-agent ramp-merging scenario.

Load-bearing premise

The load-bearing premise is that the local quadratic cost Hessians are positive semidefinite; the paper's own collision-avoidance log-barrier has an indefinite Hessian, so this premise is not met in the experiments, and if it fails the per-iteration uniqueness guarantee can collapse.

Editorial extensions

If this is right

  • Near obstacles and dense interactions, agents automatically follow the learned reference policy more closely, while in open areas they can optimize more freely.
  • Each outer iteration of the solver is a frozen linear–quadratic game with a unique Gaussian feedback Nash equilibrium, so the iterative scheme stays well-posed before convergence.
  • Cost weights inferred by the scene-aware inverse network are structurally consistent with the forward solver through the rule-based penalty, improving parameter and trajectory recovery.
  • In the reported experiments, the full adaptive method achieves zero collisions in the five-agent ramp merging scenario and lower collision counts than fixed-lambda and KKT baselines in a 20-agent scenario.

Reading between the lines

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

  • The PSD assumption gap suggests a convex surrogate for the log-barrier, or an explicit projection of $Q_{i,t}$ onto the positive semidefinite cone, would restore Proposition 1's guarantee for the same planner.
  • The state-dependent $\lambda_i$ can be read as a safety-margin proxy; a testable corollary is that collision counts should fall monotonically as $\lambda_i^{\max}$ rises in dense scenes, which the 20-agent ablation partially supports.
  • The inverse module's scene-aware weighting should be tested on naturalistic human-driving data, which the paper itself lists as future work, to separate the benefit of map context from the benefit of extra training data.
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

3 major / 3 minor

Summary. The manuscript proposes a forward–inverse dynamic game framework for multi-agent trajectory planning. The forward module iteratively linearizes nonlinear dynamics and costs around a nominal trajectory and solves a KL-regularized linear–quadratic game with a state-dependent weight λ_i(x_t), claiming a closed-form Gaussian feedback Nash equilibrium (Theorem 1), per-iteration well-posedness (Proposition 1), and Lipschitz continuity of λ_i (Proposition 2). The inverse module uses maximum-entropy IRL with a scene-aware neural network and a physics-informed rule loss to infer agent cost weights from demonstrations. Experiments are reported for cooperative navigation and ramp merging, including ablation studies, sensitivity analysis, and hardware tests.

Significance. If the theoretical results were correct, the adaptive KL-regularization scheme and the scene-aware inverse module would be a useful practical contribution: the state-dependent λ provides an intuitive mechanism to tighten reference-policy adherence near hazards, and the Lipschitz property supports stable updates. The paper is explicit about assumptions and gives proofs for Lemma 1, Theorem 1, and Propositions 1–2, and the ablation design cleanly separates the contributions of the adaptive weight and the learned cost weights. However, the central theorem contains an algebraic sign error, the main well-posedness proposition presupposes a convexity condition violated by the paper's own cost primitive, and the simulation evaluation is self-referential because the 'expert' demonstrations are generated by the same forward simulator used for rollouts. These issues are load-bearing, not cosmetic.

major comments (3)
  1. [Section IV-A, Eq. (13b) and Appendix C] The KL term in Eq. (49) expands with μ_i − μ̃_i = −P_i δx − (α_i + μ̃_i), so the linear coefficient in δx is +λ_i P_i^T(Σ̃_i)^{-1}(α_i + μ̃_i), not +λ_i P_i^T(Σ̃_i)^{-1}(α_i − μ̃_i) as printed in Eq. (13b). A one-step scalar check with A=B=Q=R=λ=Σ̃=Z_{t+1}=1, μ̃=1, and l=r=z_{t+1}=0 gives the true value coefficient z_t=1/3 by direct Bellman evaluation, whereas Eq. (13b) gives z_t=−1/3. Since z_t propagates backward into Eq. (12b), the affine feedback offsets computed by Algorithm 1 are not those of the claimed Gaussian FBNE, even under the PSD assumptions of Proposition 1.
  2. [Section IV-A, Proposition 1, and Section V-A.3, Eq. (26)] Proposition 1 is stated under Q_i_t⪰0, R_ij_t⪰0, and R_ii_t≻0 (Section IV-A, after Eq. (5)). In the experiments, the collision-avoidance primitive f_i2 = −Σ_{j≠i} log(||p_i − p_j||^2) in Eq. (26) is part of c_i_t. In two dimensions the Hessian of −log||p_i − p_j||^2 has one positive and one negative eigenvalue, so Q_i_t is indefinite at generic configurations. Therefore the premise of Proposition 1 is violated by the planner as actually implemented; the proof that Z_i_t⪰0 and Σ_i*_t≻0 does not go through, and the claimed per-iteration well-posedness and uniqueness are not established for the reported experiments. The authors need to either use a convex surrogate, project the quadraticized Hessian, or prove the relevant statements under weaker assumptions.
  3. [Section VI, Conclusion] The Conclusion states that 'expert demonstrations are currently obtained from forward simulation rather than naturalistic datasets such as INTERACTION.' Because Algorithm 2 also generates its rollout set with the same forward simulator (Lines 4–8), the quantitative metrics D_par, D_tra, and Coll measure the inverse module's ability to reproduce the simulator's own cost model, not its ability to recover genuine human objectives. The real-robot experiments are qualitative and do not report inverse-game metrics. The paper's claim of validating inference of 'unknown agents' objectives' is therefore unsupported by the presented evidence.
minor comments (3)
  1. [Eq. (13b) and Appendix C] The terminal condition is stated as z_i_T = l_i_T in Eq. (13b), but the base case in Appendix C says z_i_T = 0 provided x̄_T = x_ref; please reconcile the two conventions.
  2. [Section V-B, 4-agent CAMP] The text says Ours(fixedλ) 'reduces D_par from 1.18 (KLGame) to 1.49'; since 1.49 is larger than 1.18, the wording should be corrected or the numbers checked against Table I.
  3. [Throughout] There are several typographical errors, including 'admits admits' in Theorem 1, 'positon exchange' in Section V-A.2, and 'T rajectories' in Fig. 8; a careful proofread is needed.

Circularity Check

1 steps flagged · score 3.0 of 10

The game-theoretic derivation is independent; the only circularity is evaluative, since expert demonstrations are produced by the same forward simulator used for training and scoring.

  1. fitted input called prediction [Section VI (Conclusion); Eq. (27b); Algorithm 2 lines 4-23; Algorithm 1 line 1]
    "First, expert demonstrations are currently obtained from forward simulation rather than naturalistic datasets such as INTERACTION [33]."

    The inverse module fits ω_est by maximizing the likelihood of expert trajectories D (Eq. (23)), and Eq. (27b) scores D_tra by comparing the forward solver's reconstruction (using ω_est) with those same expert trajectories. Since the Conclusion admits that the demonstrations come from the paper's own forward simulator, the reported trajectory/parameter errors measure the inverse module's ability to invert its own generative model on its training data, not an independent prediction. This is a fitted-input-called-prediction loop confined to the evaluation: Theorem 1 and Propositions 1-2 are derived from the KL-regularized LQ game and do not inject any fitted constant into the theoretical result.

full rationale

The derivation chain is not circular. Theorem 1 is obtained by backward induction from the KL-regularized Bellman equation used in Lemma 1, with explicit first-order conditions and coefficient matching; Proposition 1 is an induction on the Riccati recursion under the stated PSD assumptions. Lemma 1 is attributed to external prior work (Lidard et al. [4]), not to the present authors, and no load-bearing uniqueness theorem is imported from the authors' own publications. The reference-policy mean in Eq. (8a) is an explicit design choice, not a hidden ansatz smuggled via citation. The only genuine circularity is the empirical loop: the Conclusion acknowledges that expert demonstrations were generated by forward simulation, while the inverse solver both trains on and is evaluated against those same demonstrations (Eqs. (23) and (27b)). That makes the experimental validation a self-consistency check rather than an external benchmark. The reported algebraic sign issue in Eq. (13b) is a correctness defect in the Riccati recursion, not a circular-reasoning pattern, so it is excluded from the circularity score.

Assumptions & free parameters 5 free parameters · 4 assumptions · 0 invented entities

The framework introduces no new physical entities. The main loaded assumptions are the convexity of the quadraticized cost, which the paper's own log-barrier violates, and the nonsingularity of the coupled Riccati systems. The lambda parameters, sigma, and nu_rule are hand-chosen hyperparameters rather than learned quantities.

free parameters (5)
  • lambda_min = 0.5
    Lower bound of the state-dependent KL weight, set by hand in Section V-A.4 for all scenarios.
  • lambda_max = 5.0
    Upper bound of the KL weight, set by hand.
  • sigma = 1.0
    Gaussian kernel width in Eq. (15), set by hand.
  • nu_rule = 0.1
    Weight of the rule-based loss in Eq. (22), chosen by hand after a sensitivity sweep.
  • tau = not specified
    Temperature in the reference Boltzmann policy Eq. (6a); no value is given in the text.
assumptions (4)
  • ad hoc to paper The local quadratic cost Hessians satisfy Q_i >= 0, R_ij >= 0, R_ii > 0.
    Invoked in Theorem 1 and Proposition 1 to prove well-posedness; violated by the log-barrier cost in Eq. (26).
  • domain assumption The coupled linear systems in Eqs. (12a)-(12b) are nonsingular at every time step (Assumption A4).
    Used in Proposition 1 to assert uniqueness of the feedback Nash equilibrium; assumed, not proven.
  • domain assumption The reference policy Laplace approximation yields a strictly positive definite covariance (Eq. (8b)).
    Needed for Gaussian closure and for Sigma~ > 0 in Proposition 1.
  • domain assumption The nominal trajectory remains collision-free (Assumption A2).
    Guarantees positive d_obs in Eq. (15) and bounded lambda; not guaranteed by Algorithm 1.

how reviews work

0 comments
Cite this review

Pith. "Pith review of A Forward-Inverse Dynamic Game Framework for Enhanced Multi-Agent Trajectory Planning." pith.science (2026). https://pith.science/paper/FFZKZTZA

@misc{pith2026260801636,
  author       = {Pith},
  title        = {Pith review of: A Forward-Inverse Dynamic Game Framework for Enhanced Multi-Agent Trajectory Planning},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/FFZKZTZA}},
  note         = {Machine review of arXiv:2608.01636}
}
read the original abstract

This paper studies feedback Nash equilibrium (FBNE) seeking for multi-agent trajectory planning in nonlinear dynamical systems with unknown agents' objectives and state-dependent inter-agent coupling. While dynamic game theory provides a principled framework for such problems, existing approaches typically assume fully rational agents with known objectives or rely on fixed regularization, limiting their ability to capture bounded rationality and spatially varying interaction intensity in safety-critical settings. To this end, we propose a KL-regularized dynamic game with a state-dependent weight that adaptively balances optimality and behavioral priors. To infer unknown cost parameters from demonstrated behaviors, we develop a context-aware inverse game module based on maximum-entropy inverse reinforcement learning with physics-informed regularization, ensuring structural consistency with the forward game. We establish per-iteration well-posedness of the regularized local game and show that the adaptive weighting function remains Lipschitz continuous under bounded nominal-trajectory updates. Numerical simulations and multi-robot experiments on cooperative navigation and merging scenarios validate the effectiveness of the proposed framework.

Figures

Figures reproduced from arXiv: 2608.01636 by the authors.

Figure 1
Figure 1. Red ego vehicle merging from an on-ramp onto a highway among four surrounding agents with uncertain intentions. expert demonstrations for more realistic decisions. We further introduce an adaptive weighting scheme to bal￾ance data-driven influence, improving accuracy and con￾textual adaptability in trajectory planning. • We introduce a scene-aware neural network that is ca￾pable of jointly encoding multi-agent traje… view at source ↗
Figure 2
Figure 2. The proposed forward–inverse game framework. (1) In the inverse game, agent goals are inferred from expert data. (2) In the forward game, KL divergence is incorporated into the formulation, and the feedback Nash equilibrium under nonlinear dynamics is solved via iterative linearization. k we apply a Laplace approximation [28] to π˜ i . Specifically, we perform a second-order Taylor expansion of Qi , equiva￾lently of… view at source ↗
Figure 3
Figure 3. Architecture of the scene-aware neural network [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: CAMP: (a) Four agents perform pairwise position exchange. (b) Two agents perform the same exchange task [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]
Figure 5
Figure 5. Figure 5: RM: Agent2 completes a collision-free merge. Agent4 and 5 are excluded from the interaction due to lane separation but included for metric evaluation. 2) Ablation Study and Comparison with Baselines: We con￾ducted 50 independent trials under identical settings. To dise…
Figure 6
Figure 6. Figure 6: Comparison of trajectory and parameter errors. C. Sensitivity Analysis of νrule To evaluate the sensitivity of the method to the hyperpa￾rameter νrule in Eq. (22), we perform an ablation study in the RM scenario by varying νrule ∈ {0, 0.01, 0.1, 1.0, 10.0} while keepin…
Figure 7
Figure 7. Figure 7: State-dependent λi convergence in the RM (5 agents) scenario. warm-starting and a maximum of kmax = 15 iterations, the average per-step planning time was 35 ms (≈28 Hz), satisfying the 10 Hz real-time requirement imposed by ∆t = 0.1 s. Z [PITH_FULL_IMAGE:figures/full_…
Figure 8
Figure 8. Figure 8: Trajectories executed using learned strategies in position exchange (top) and trajectory crossing (bottom). VI. CONCLUSION We proposed a forward–inverse dynamic game framework for multi-agent trajectory planning that couples scene-aware cost inference with a KL-regular…

Review history (2 revisions) →

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

34 extracted references · 29 canonical work pages

  1. [1]

    Predictability awareness for efficient and robust multi-agent coordination,

    R. C. Gil, D. J. Ornia, K. A. Mustafa, and J. Alonso Mora, “Predictability awareness for efficient and robust multi-agent coordination,” inProceed- ings of the 24th International Conference on Autonomous Agents and Multiagent Systems, 2025, pp. 886–894

    work page 2025

  2. [2]

    Learning to play trajectory games against opponents with unknown objectives,

    X. Liu, L. Peters, and J. Alonso-Mora, “Learning to play trajectory games against opponents with unknown objectives,”IEEE Robotics and Automation Letters, vol. 8, no. 7, pp. 4139–4146, 2023

    work page 2023

  3. [3]

    Cost inference for feedback dynamic games from noisy partial state observations and incomplete trajectories,

    J. Li, C.-Y . Chiu, L. Peters, S. Sojoudi, C. Tomlin, and D. Fridovich- Keil, “Cost inference for feedback dynamic games from noisy partial state observations and incomplete trajectories,” inProceedings of the 2023 International Conference on Autonomous Agents and Multiagent Systems, 2023, pp. 1062–1070

    work page 2023

  4. [4]

    Blending data-driven priors in dynamic games,

    J. Lidard, H. Hu, A. Hancock, Z. Zhang, A. G. Contreras, V . Modi, J. DeCastro, D. Gopinath, G. Rosman, and N. E. Leonard, “Blending data-driven priors in dynamic games,” inProceedings of Robotics: Science and Systems (RSS), Delft, Netherlands, Jul. 2024

    work page 2024

  5. [5]

    Open-loop and feedback nash trajectories for competitive racing with ilqgames,

    M. Rowold, A. Langmann, B. Lohmann, and J. Betz, “Open-loop and feedback nash trajectories for competitive racing with ilqgames,” in 2024 IEEE 27th International Conference on Intelligent Transportation Systems (ITSC). IEEE, 2024, pp. 1827–1834

    work page 2024

  6. [6]

    A game of social forces: Integrating non-cooperative game theory with social force model for a socially-acceptable mobile robot navigation,

    G. Galati, A. Usai, G. Vignolo, S. Macr `ı, and A. Rizzo, “A game of social forces: Integrating non-cooperative game theory with social force model for a socially-acceptable mobile robot navigation,”Journal of Intelligent & Robotic Systems, vol. 111, no. 4, p. 110, 2025

    work page 2025

  7. [7]

    Online and offline learning of player objectives from partial observations in dynamic games,

    L. Peters, V . Rubies-Royo, C. J. Tomlin, L. Ferranti, J. Alonso-Mora, C. Stachniss, and D. Fridovich-Keil, “Online and offline learning of player objectives from partial observations in dynamic games,”The International Journal of Robotics Research, vol. 42, no. 10, pp. 917–937, 2023

    work page 2023

  8. [8]

    Neural-network-based dis- tributed generalized nash equilibrium seeking for uncertain nonlinear multiagent systems,

    W. Huo, L. Huang, S. Dey, and L. Shi, “Neural-network-based dis- tributed generalized nash equilibrium seeking for uncertain nonlinear multiagent systems,”IEEE Transactions on Control of Network Systems, vol. 11, no. 3, pp. 1323–1334, 2023

    work page 2023

Show all 34 references
  1. [9]

    Continuous-time distributed generalized nash equilibrium seeking in nonsmooth fuzzy aggregative games,

    J. Liu, X. Liao, J.-S. Dong, and A. Mansoori, “Continuous-time distributed generalized nash equilibrium seeking in nonsmooth fuzzy aggregative games,”IEEE Transactions on Control of Network Systems, vol. 11, no. 3, pp. 1262–1274, 2023

    2023

  2. [10]

    Fully distributed primal-dual generalized nash equilibrium seeking algorithm under partial-decision information,

    X. Liao, W. Meng, and X. Li, “Fully distributed primal-dual generalized nash equilibrium seeking algorithm under partial-decision information,” IEEE Transactions on Control of Network Systems, 2025

    2025

  3. [11]

    Distributed nash equilibrium seeking for games in uncertain nonlinear systems via adaptive backstepping approach,

    Q. Meng and Q. Ma, “Distributed nash equilibrium seeking for games in uncertain nonlinear systems via adaptive backstepping approach,”IEEE Transactions on Control of Network Systems, vol. 12, no. 1, pp. 1188– 1198, 2024

    2024

  4. [12]

    C ¸ inlar,Probability and Stochastics

    E. C ¸ inlar,Probability and Stochastics. Springer, 2011

    2011

  5. [13]

    Lucidgames: Online unscented inverse dynamic games for adaptive trajectory prediction and planning,

    S. Le Cleac’h, M. Schwager, and Z. Manchester, “Lucidgames: Online unscented inverse dynamic games for adaptive trajectory prediction and planning,”IEEE Robotics and Automation Letters, vol. 6, no. 3, pp. 5485–5492, 2021

    2021

  6. [14]

    Maximum-entropy multi-agent dynamic games: Forward and inverse solutions,

    N. Mehr, M. Wang, M. Bhatt, and M. Schwager, “Maximum-entropy multi-agent dynamic games: Forward and inverse solutions,”IEEE Transactions on Robotics, vol. 39, no. 3, pp. 1801–1815, 2023

    2023

  7. [15]

    Multiagent graphical games with inverse reinforcement learning,

    V . S. Donge, B. Lian, F. L. Lewis, and A. Davoudi, “Multiagent graphical games with inverse reinforcement learning,”IEEE Transactions on Control of Network Systems, vol. 10, no. 2, pp. 841–852, 2022

    2022

  8. [16]

    Nonlinear programming,

    H. W. Kuhn and A. W. Tucker, “Nonlinear programming,” inProceed- ings of the Second Berkeley Symposium on Mathematical Statistics and Probability. University of California Press, Berkeley and Los Angeles, 1951, pp. 481–492

    1951

  9. [17]

    Hierarchical game-theoretic planning for autonomous vehicles,

    J. F. Fisac, E. Bronstein, E. Stefansson, D. Sadigh, S. S. Sastry, and A. D. Dragan, “Hierarchical game-theoretic planning for autonomous vehicles,” in2019 International Conference on Robotics and Automation (ICRA). IEEE, 2019, pp. 9590–9596

    2019

  10. [18]

    Mpogames: Efficient multimodal partially observable dy- namic games,

    O. So, P. Drews, T. Balch, V . Dimitrov, G. Rosman, and E. A. Theodorou, “Mpogames: Efficient multimodal partially observable dy- namic games,” in2023 IEEE International Conference on Robotics and Automation (ICRA). IEEE, 2023, pp. 3189–3196

    2023

  11. [19]

    Human-like robot action policy through game-theoretic intent inference for human–robot collaboration,

    Y . Sheng, Y . Wang, H. Cheng, H. Zhao, and H. Ding, “Human-like robot action policy through game-theoretic intent inference for human–robot collaboration,”IEEE Transactions on Robotics, vol. 41, pp. 5411–5430, 2025

    2025

  12. [20]

    Bas ¸ar and G

    T. Bas ¸ar and G. J. Olsder,Dynamic noncooperative game theory. SIAM, 1998

    1998

  13. [21]

    The computa- tion of approximate generalized feedback nash equilibria,

    F. Laine, D. Fridovich-Keil, C.-Y . Chiu, and C. Tomlin, “The computa- tion of approximate generalized feedback nash equilibria,”SIAM Journal on Optimization, vol. 33, no. 1, pp. 294–318, 2023

    2023

  14. [22]

    Factorised active inference for strategic multi-agent interactions,

    J. Ruiz-Serra, P. Sweeney, and M. S. Harr ´e, “Factorised active inference for strategic multi-agent interactions,” inProceedings of the 24th In- ternational Conference on Autonomous Agents and Multiagent Systems, 2025, pp. 1793–1802

    2025

  15. [23]

    Inverse differential games with mixed inequality constraints,

    C. Awasthi and A. Lamperski, “Inverse differential games with mixed inequality constraints,” in2020 American Control Conference (ACC), Jul 2020, p. 2182–2187

    2020

  16. [24]

    Guided cost learning: deep in- verse optimal control via policy optimization,

    C. Finn, S. Levine, and P. Abbeel, “Guided cost learning: deep in- verse optimal control via policy optimization,” inProceedings of the 33rd International Conference on International Conference on Machine Learning-Volume 48, 2016, pp. 49–58

    2016

  17. [25]

    Multi-agent adversarial inverse reinforce- ment learning,

    L. Yu, J. Song, and S. Ermon, “Multi-agent adversarial inverse reinforce- ment learning,”International Conference on Machine Learning, May 2019

    2019

  18. [26]

    Efficient iterative linear-quadratic approximations for nonlinear multi- player general-sum differential games,

    D. Fridovich-Keil, E. Ratner, L. Peters, A. D. Dragan, and C. J. Tomlin, “Efficient iterative linear-quadratic approximations for nonlinear multi- player general-sum differential games,” in2020 IEEE International Conference on Robotics and Automation (ICRA), 2020, pp. 1475–1481

    2020

  19. [27]

    Finite-time analysis of minimax q-learning,

    N. Jeong and D. Lee, “Finite-time analysis of minimax q-learning,” in Reinforcement Learning Conference, 2025

    2025

  20. [28]

    C. M. Bishop and N. M. Nasrabadi,Pattern recognition and machine learning. Springer, 2006, vol. 4, no. 4

    2006

  21. [29]

    Maximum entropy deep inverse reinforcement learning,

    M. Wulfmeier, P. Ondruska, and I. Posner, “Maximum entropy deep inverse reinforcement learning,”arXiv preprint:1507.04888, 2015

    2015 arXiv

  22. [30]

    Vectornet: Encoding hd maps and agent dynamics from vectorized rep- resentation,

    J. Gao, C. Sun, H. Zhao, Y . Shen, D. Anguelov, C. Li, and C. Schmid, “Vectornet: Encoding hd maps and agent dynamics from vectorized rep- resentation,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020, pp. 11 525–11 533

    2020

  23. [31]

    Maximum entropy inverse reinforcement learning

    B. D. Ziebart, A. L. Maas, J. A. Bagnell, A. K. Deyet al., “Maximum entropy inverse reinforcement learning.” inAaai, vol. 8. Chicago, IL, USA, 2008, pp. 1433–1438

    2008

  24. [32]

    Combining reinforcement learning with model predictive control for on-ramp merging,

    J. Lubars, H. Gupta, S. Chinchali, L. Li, A. Raja, R. Srikant, and X. Wu, “Combining reinforcement learning with model predictive control for on-ramp merging,” in2021 IEEE International Intelligent Transporta- tion Systems Conference (ITSC). IEEE, 2021, pp. 942–947

    2021

  25. [33]

    INTERACTION dataset: An INternational, Adversarial and Cooperative moTion dataset in interactive driving scenarios with semantic maps,

    W. Zhan, L. Sun, D. Wang, H. Shi, A. Clausse, M. Naumann, J. K ¨ummerle, H. K ¨onigshof, C. Stiller, A. de La Fortelle, and M. Tomizuka, “INTERACTION dataset: An INternational, Adversarial and Cooperative moTion dataset in interactive driving scenarios with semantic maps,”arXi...

    1910 arXiv

  26. [34]

    Rudin,Principles of Mathematical Analysis, 3rd ed

    W. Rudin,Principles of Mathematical Analysis, 3rd ed. McGraw-Hill, 1976

    1976

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.