Pith. sign in

REVIEW 5 major objections 8 minor 44 references

Learning for Long-Horizon Planning via Neuro-Symbolic Abductive Imitation

T0 review · 5 major / 8 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read ABIL uses abductive reasoning over a symbolic knowledge base to generate predicate labels from unannotated demonstrations, then selects per-operator imitation policies through symbolic planning, yielding better data efficiency and…

desk verdict The core idea is real and the gains are big, but the zero-shot result is likely pre-encoded in the supplied KB, so read the strong claims with caution. read the letter →

arxiv 2411.18201 v1 pith:SYHME6SM submitted 2024-11-27 cs.LG cs.AI

Jie-Jing Shao , Hao-Ran Hao , Xiao-Wen Yang , Yu-Feng Li This is my paper · ORCID
classification cs.LGcs.AI
keywords ImitationLearningAbductiveNeuro-SymbolicEmbodiedArtificialIntelligenceLong-HorizonPlanningSymbolicDataEfficiency
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 proposes ABIL, a framework that blends imitation learning with symbolic abductive reasoning so agents can solve long-horizon tasks from raw observations without any predicate-level annotations. ABIL starts from a human-supplied knowledge base—a state machine whose nodes are sub-task conditions and whose edges are logical operators—and uses abductive reasoning to infer which symbolic states each demonstration must have passed through, turning those inferences into pseudo-labels that train a perception module. The learned perception maps observations to symbols, a symbolic planner picks the next logical operator, and an ensemble of behavior policies, each trained for one operator, executes the low-level actions. In experiments on BabyAI, Mini-BEHAVIOR, and CLIPort, ABIL reports substantially higher data efficiency, stronger generalization, and zero-shot transfer to composed tasks than behavior cloning, Decision Transformer, and PDSketch. If the claim holds, imitation learning would no longer need ground-truth predicate annotations to benefit from symbolic planning's long-horizon robustness.

What carries the argument

The load-bearing object is the knowledge base $G$, a directed graph state machine whose vertices are sets of ground atoms (sub-task conditions) and whose edges carry a logical operator $op$ with add/delete effects. From this graph ABIL derives 'sequential consistency' abduction rules—task completed, $a$ before $b$, either $a$ or $b$, and both $a$ and $b$ in any order—which turn an unlabeled trajectory into a constrained symbolic sequence. That sequence solves the joint optimization in Eq. 1, providing pseudo-labels for predicate classifiers $f_p$ at the object level. The same graph supports a symbolic plan, and Eq. 2 maps each perceived symbolic state to the current operator, selecting which imitation policy $h_{op}$ executes next. These four pieces—KB as state machine, sequential-consistency abduction, object-level predicate grounding, and operator-ensemble selection—together carry the argument for data-efficient and generalizable long-horizon imitation.

What would settle it

Take a BabyAI task family and construct two versions of the knowledge base, one correct and one with the sequence of Pick and Open reversed, then train ABIL on identical demonstrations. If success rates and predicate accuracy stay essentially unchanged, the abductive sequential-consistency constraints are not doing the claimed labeling work; if they collapse, the knowledge-base dependence is confirmed.

Watch

Extended reading notes

Core claim

ABIL's central claim is that the hidden symbolic structure of expert demonstrations can be recovered by abduction and then used to make imitation learning compositional. Given a knowledge base $G$ and a trajectory of raw observations, the framework alternates between a perception network $f$ and a symbolic sequence $\{z_t\}$: Eq. 1 chooses the $z_t$ that are simultaneously close to $f(s_t)$ and consistent with $G$ under the 'sequential consistency' principle, and those $z_t$ are the pseudo-labels that train $f$. Each predicate is grounded by an object-level classifier, so the perception module learns to detect task-relevant symbols without human annotations. Once observations map to symbolic states, Eq. 2 selects the logical operator $op_t$ from the symbolic plan for the current state, and Eq. 4 trains one behavior policy per operator by imitation, with the operator's arguments coming from Eq. 3. The authors argue that this operator-level decomposition is what lets the agent handle long horizons, novel object counts, unseen distractors, and composed tasks, because reasoning and imitation happen at the level of abstract operators rather than raw action sequences.

Load-bearing premise

The load-bearing premise is that the expert-supplied knowledge base is accurate and sufficient for every demonstration in the task family; if the knowledge base is wrong, incomplete, or inconsistent with what the expert actually does, the abduced pseudo-labels and the operator selections built on them are wrong too.

Editorial extensions

If this is right

  • Without any predicate annotations, an agent can be trained from raw demonstrations and still ground the task-relevant symbols that symbolic planning needs.
  • Much less data is needed for reliable neuro-symbolic grounding: in BabyAI, ABIL reaches near-perfect predicate accuracy with less than 20% of the demonstrations PDSketch uses.
  • Long-horizon tasks that break model-based planners become tractable, including Mini-BEHAVIOR tasks of roughly 45 to 106 decision steps where PDSketch times out.
  • Agents can zero-shot compose learned operators, as when policies trained on pickup/open or on throwing one leftover transfer to the unlock task or to throwing two or three leftovers.
  • Inference stays close to learning-method speed while gaining symbolic-planning generality, because search happens at the operator level rather than in raw observation-action space.

Reading between the lines

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

  • Beyond the paper, a natural extension is to let a vision-language model propose predicate candidates instead of relying on the hand-built knowledge base, which would make ABIL's abductive labeling loop semi-automatic.
  • ABIL's dependence on $G$ suggests a diagnostic use: demonstrations that cannot be mapped onto $G$ could be flagged automatically, turning knowledge-base incompleteness into a detectable signal for acquiring new operators.
  • The paper evaluates at most a handful of operators per task; whether the one-policy-per-operator scheme and the graph planner scale to hundreds of operators is an untested question that a larger benchmark could answer.
  • The paper's determinism and full-observability assumptions point to a belief-space version of ABIL; if that extension is attempted, the abduction constraints would need to be reformulated for distributional symbolic states.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

5 major / 8 minor

Summary. The paper proposes ABIL, a neuro-symbolic imitation learning framework for long-horizon goal-based planning. The method assumes a symbolic knowledge base G (a finite-state machine over sub-task nodes) for each task, plus expert demonstrations without predicate annotations. ABIL operates in two stages: (1) abductive pseudo-labeling, where a perception network f is trained by iteratively proposing symbolic state labels z for each demonstration frame that satisfy G while staying close to f's current predictions (Eq. 1), using 'sequential consistency' rules for sequential, alternative, and conjunctive sub-task structures (Section 3.2); and (2) symbolic-grounded imitation, where a separate behavior module is trained per logical operator and the operator is selected at inference by matching the perceived state to plan nodes (Eq. 2), yielding a gated, reasoning-aware policy (Section 3.3). The paper evaluates ABIL on BabyAI, Mini-BEHAVIOR, and CLIPort against BC, DT, and PDSketch, reporting success rates, neuro-symbolic grounding accuracy under varying data budgets, data-efficiency curves, and zero-shot transfer experiments for composed or repeated tasks. The main claims are significantly improved data efficiency and generalization relative to the baselines, achieved without predicate annotations.

Significance. The core idea—replacing predicate-level annotation with an abduction loop constrained by a task-level state machine, and gating per-operator behavior modules by symbolic state—is well motivated and timely. If the results hold, the practical significance is real: per-operator decomposition plus symbolic gating yields large gains on genuinely long-horizon tasks (Mini-BEHAVIOR tasks of 45-106 steps; CLIPort assembly tasks, e.g., Assembling-kits from 0.187 to 0.829), and the grounding-accuracy curves (Figures 4, 6) show a data-efficiency advantage over PDSketch that is the kind of falsifiable claim this community values. The manuscript also ships code (Appendix C), uses 100 evaluations and three seeds throughout, compares against the strong neuro-symbolic baseline PDSketch, includes an explicit robustness probe with imperfect grounding (Table 6), and states its limitations honestly (Section 5).

major comments (5)
  1. [Section 4.5; Appendix D.2] The zero-shot transfer results in Section 4.5 do not support the claimed compositional generalization because the target composition is already present in the supplied knowledge base. In the Throwing-away-leftovers experiment, training uses one hamburger, but the knowledge base in Appendix D.2 explicitly enumerates Pick(hamburger_1, countertop), Pick(hamburger_2, countertop), and Pick(hamburger_3, countertop) as well as Place(hamburger_1, ashcan), Place(hamburger_2, ashcan), and Place(hamburger_3, ashcan) as nodes and edges of G. If this graph is the G used at test time, Eq. (2) reads the repeated-task structure directly off G, so the zero-shot margins in Table 4 (ABIL at 0.49-0.70 vs. 0.05-0.18 for BC/DT on Throw 2 and Throw 3) reflect privileged input structure rather than learned generalization. The manuscript must state whether G is a template over parametric object variables (in which case the instantiation procedure for 1, 2, and 3 objects must be described) or an object-specific plan (in which case the zero-shot claim in Table 4 is confounded and should be withdrawn or re-run with a G that does not enumerate the test objects).
  2. [Section 4, Tables 1-4; Sections 3.2-3.3] The headline comparisons against BC and DT are confounded by the privileged per-task knowledge base G. In BabyAI, Appendix D.1 gives each task a state machine that already encodes the subgoal order (e.g., the Unlock graph contains Pick(red_key) before Goto(red_door) and Open(red_door)), and ABIL's operator selection (Eq. 2) and per-operator behavior modules (Eq. 4) consume exactly this structure, while BC and DT receive only raw observations and actions. Consequently, the large margins in Tables 1-4 could be driven by the operator decomposition itself rather than by the abductive pseudo-labeling that the paper emphasizes. To support the attribution in the abstract, the paper should add ablations that (i) provide the same operator decomposition to the baselines (e.g., per-operator BC/DT trained with ground-truth operator labels and an oracle selector derived from G) and (ii) replace the abduced labels with labels read off G directly, isolating the perception component. The comparison with PDSketch is less affected because PDSketch's full mode also consumes the symbolic transition structure; the confound specifically concerns the BC/DT baselines, which are the primary comparators for the data-efficiency and generalization claims.
  3. [Section 3.2, Eq. (1)] The formal abduction objective and the implemented procedure are not the same object. Eq. (1) defines the pseudo-labels as the argmin of a constrained optimization over the whole trajectory, but the bullet list that follows gives four informal consistency rules ('task a before task b', 'either a or b', and so on), and the paper never states the algorithm that solves Eq. (1), nor shows that the sequential-abduction rules are equivalent to (or an approximation of) that argmin. Since the perception network f is trained against these pseudo-labels, the correctness of the entire perceptual component depends on this step; the paper should specify the exact inference procedure (e.g., a dynamic program over the state-machine graph) and state any optimality gaps explicitly.
  4. [Abstract; Section 3.3; Section 4] The paper claims 'significantly improved data efficiency and generalization across various long-horizon tasks' (abstract) and describes 'a policy ensemble' (Section 3.3), but neither claim is fully supported by the reported experiments. No significance tests are reported in Section 4, and several table cells contradict the blanket claim: in Table 2, ABIL-DT is below BC on Cleaning A Car Basic (0.340 vs. 0.417) and below BC on Collect misplaced items Gen (0.279 vs. 0.325), with overlapping error bars in several other rows (e.g., Cleaning A Car Gen). Regarding the ensemble, the experiments evaluate only ABIL-BC and ABIL-DT as two separate single-backbone configurations; there is no experiment in which multiple behavior modules are combined, so the 'ensemble' terminology in the abstract and Section 3.3 overstates the method as evaluated. The authors should either add significance tests and temper the per-task claims, or add an actual ensemble evaluation.
  5. [Table 2] Table 2 is internally inconsistent as presented. For several tasks (Making tea, Moving boxes to storage, Cleaning A Car, Throwing away leftovers, Putting away dishes, Sorting books, Laying wood floors, Watering houseplants, Cleaning shoes, Collect misplaced items, Organizing file cabinet), the Basic evaluation lists PDSketch as '> 5 minutes', yet the corresponding Gen rows either print numeric PDSketch values despite the Basic timeout (Making tea Gen shows 0.074±0.053) or appear to omit the PDSketch column entirely (Moving boxes Gen, Cleaning A Car Gen, and others list only four values for five methods). The authors should fix the table alignment and state explicitly whether a timed-out PDSketch run is counted as a failure (0%) or excluded, since the comparison in Section 4.2 hinges on this protocol.
minor comments (8)
  1. [Abstract; Section 1] The abstract and introduction state that ABIL 'generates predicate candidates' for perception, but Section 3.1 takes the predicate set P as given and Section 3.2 generates only pseudo-labels (abduced symbolic states) over a fixed predicate vocabulary; please reword to avoid implying predicate invention, or describe the candidate-generation mechanism if one exists.
  2. [Section 3.3, Eq. (2)] The operator selection rule is defined by an existential condition ('exists k in [0,K)'), so if the perceived state satisfies the node conditions of multiple plan nodes simultaneously, op_t is not uniquely determined; the paper should state the deterministic selection rule used (e.g., first matching node in plan order) and confirm that it was followed in the experiments.
  3. [Figures 4 and 6] The grounding-accuracy curves are plotted without error bars or standard deviations even though the text reports three seeds; adding them would let readers assess the 'less than 20% of their data' claim in Section 4.1 against the spread.
  4. [Figures 4 and 6] The paper does not state whether the grounding accuracy is measured on held-out demonstrations or on the same demonstrations used to generate the pseudo-labels; since the pseudo-labels are derived from the training demonstrations themselves, please specify the evaluation protocol to rule out a self-consistency measure.
  5. [Section 4.1] The quantitative claim 'requiring less than 20% of their data' is not tied to a specific curve or threshold in Figure 4; please state the exact demonstration budgets and accuracy levels that support the 20% figure.
  6. [Appendix B.1, Table 6] In Table 6, ABIL-DT Pickup Basic at 90% grounding accuracy (0.860±0.017) is higher than at 100% (0.845±0.035), and at 50% accuracy two cells fall below the data-driven baseline (ABIL-BC Pickup Basic 0.717 vs. 0.723; ABIL-DT Putting-blocks-in-bowls 0.462 vs. 0.539); the robustness conclusion in B.1 should be qualified to the accuracy regime where it actually holds.
  7. [Section 5] The limitation statement (2) concedes dependence on an accurate and sufficient knowledge base, but the paper never tests robustness to KB errors (as opposed to perception errors in Table 6); a short experiment with corrupted or incomplete G would substantially strengthen the honest framing given in the conclusion.
  8. [References] References [21] and [28] lack venue information (a CoRR preprint and an unpublished manuscript, respectively); please complete the entries before camera-ready.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: ABIL's inputs (knowledge base G) are explicit, and the learned components are evaluated against external goals and ground-truth labels; the zero-shot KB-instantiation ambiguity is a confound, not a circular derivation.

full rationale

The derivation chain is self-contained with respect to the paper's own equations. ABIL takes a user-supplied symbolic knowledge base G, uses Eq. 1 to abduce pseudo-labels for the perception module f, and uses Eq. 2 to select per-operator policies h_op. The central empirical claims are evaluated against held-out ground-truth predicate labels and external task success, not against the constraints used to generate the pseudo-labels, so the 'predictions' are not forced by the fitting procedure. G is an explicit input, not an output of the method; the neural components are evaluated on external task success and external predicate accuracy. The paper's own limitation statement in Section 5 concedes that ABIL 'relies on an accurate and sufficient knowledge base,' which is an input assumption rather than a circular step. Minor self-citations (e.g., Refs. [27], [28], [38], [39]) appear in related work and future-work discussions and are not load-bearing; no uniqueness theorem is imported from the authors' prior work. One caveat worth flagging: the zero-shot Throwing-away-leftovers experiment (Section 4.5) trains on 1 hamburger and tests on 2 or 3, while Appendix D.2's illustrative knowledge-base graph contains hamburger_1, hamburger_2, and hamburger_3 nodes. If that graph were the actual G used during training, the test plan would be pre-encoded in the input. The paper does not specify how G is instantiated for variable object counts, which is a potential confound in the zero-shot claim, but this is not circularity in the derivation chain because G is an input, not a quantity derived from ABIL's own outputs, and the final success still requires the learned perception and behavior modules to execute the plan.

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

The central claim rests on the provided knowledge base and on the feasibility of abductive pseudo-labeling; neither is derived within the paper. No new physical or logical entities are introduced.

assumptions (4)
  • domain assumption The environment is deterministic and fully observable.
    Stated in Section 5 limitations: 'The current framework assumes deterministic and fully observable environments, consistent with existing work [24, 30, 31].' This underpins the state-machine transition model and abduction constraints.
  • domain assumption An accurate and sufficient symbolic knowledge base G is available for each task family.
    Acknowledged in Section 5: 'ABIL assumes the availability of a symbolic solution and relies on an accurate and sufficient knowledge base.' The entire abduction and operator selection use G.
  • ad hoc to paper The abduction optimization in Equation 1 yields a feasible and informative pseudo-label assignment.
    The paper does not analyze identifiability or convergence of the sequential consistency constraints; it is assumed that the pseudo-labels are consistent enough to train the perception network.
  • domain assumption Perception operates at object level with reliable object detection or feature extraction.
    Section 3.2: 'we train the perception module f at the object level... object-level features could be obtained via an object detection model in practice.' This assumes that object detection succeeds in the environments.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Learning for Long-Horizon Planning via Neuro-Symbolic Abductive Imitation." pith.science (2026). https://pith.science/paper/SYHME6SM

@misc{pith2026241118201,
  author       = {Pith},
  title        = {Pith review of: Learning for Long-Horizon Planning via Neuro-Symbolic Abductive Imitation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/SYHME6SM}},
  note         = {Machine review of arXiv:2411.18201}
}
read the original abstract

Recent learning-to-imitation methods have shown promising results in planning via imitating within the observation-action space. However, their ability in open environments remains constrained, particularly in long-horizon tasks. In contrast, traditional symbolic planning excels in long-horizon tasks through logical reasoning over human-defined symbolic spaces but struggles to handle observations beyond symbolic states, such as high-dimensional visual inputs encountered in real-world scenarios. In this work, we draw inspiration from abductive learning and introduce a novel framework \textbf{AB}ductive \textbf{I}mitation \textbf{L}earning (ABIL) that integrates the benefits of data-driven learning and symbolic-based reasoning, enabling long-horizon planning. Specifically, we employ abductive reasoning to understand the demonstrations in symbolic space and design the principles of sequential consistency to resolve the conflicts between perception and reasoning. ABIL generates predicate candidates to facilitate the perception from raw observations to symbolic space without laborious predicate annotations, providing a groundwork for symbolic planning. With the symbolic understanding, we further develop a policy ensemble whose base policies are built with different logical objectives and managed through symbolic reasoning. Experiments show that our proposal successfully understands the observations with the task-relevant symbolics to assist the imitation learning. Importantly, ABIL demonstrates significantly improved data efficiency and generalization across various long-horizon tasks, highlighting it as a promising solution for long-horizon planning. Project website: \url{https://www.lamda.nju.edu.cn/shaojj/KDD25_ABIL/}.

Figures

Figures reproduced from arXiv: 2411.18201 by the authors.

Figure 1
Figure 1. Our framework, ABductive Imitation Learning (ABIL), achieves neuro-symbolic grounding of imitation in complex [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Illustration of the Knowledge Base. annotations to help imitation learning. The most related work to ours is PDSketch [24], which utilizes neural networks as the basic modules of human-specified programming structures and learns an object-factored transition model that supports generic neural￾network-based representations for predicates and action effects. However, we find that its planning, based on the raw-observa… view at source ↗
Figure 3
Figure 3. The Framework of our Abductive Imitation Learning. [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Accuracy of neuro-symbolic grounding under varying data budgets. [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 5
Figure 5. Figure 5: Results under varying data budgets [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]
Figure 6
Figure 6. Figure 6: Accuracy of neuro-symbolic grounding under [PITH_FULL_IMAGE:figures/full_fig_p008_6.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

44 extracted references · 43 canonical work pages

  1. [1]

    Michael Bain and Claude Sammut. 1995. A Framework for Behavioural Cloning. In Machine Intelligence

    work page 1995

  2. [2]

    Bhattacharyya, Blake Wulfe, Derek J

    Raunak P. Bhattacharyya, Blake Wulfe, Derek J. Phillips, Alex Kuefler, Jeremy Morton, Ransalu Senanayake, and Mykel J. Kochenderfer. 2023. Modeling Human Driving Behavior Through Generative Adversarial Imitation Learning. IEEE Transactions on Intelligent Transportation Systems 24, 3 (2023), 2874–2887

    work page 2023

  3. [3]

    Tom B. Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah, Jared Kaplan, Prafulla Dhariwal, Arvind Neelakantan, Pranav Shyam, Girish Sastry, Amanda Askell, Sandhini Agarwal, Ariel Herbert-Voss, Gretchen Krueger, Tom Henighan, Rewon Child, Aditya Ramesh, Daniel M. Ziegler, Jeffrey Wu, Clemens Winter, Christopher Hesse, Mark Chen, Eric Sigler, Mateusz Litwin...

    work page 2020

  4. [4]

    Lili Chen, Kevin Lu, Aravind Rajeswaran, Kimin Lee, Aditya Grover, Misha Laskin, Pieter Abbeel, Aravind Srinivas, and Igor Mordatch. 2021. Decision Transformer: Reinforcement Learning via Sequence Modeling. In Advances in Neural Information Processing Systems 34 . 15084–15097

    work page 2021

  5. [5]

    Maxime Chevalier-Boisvert, Dzmitry Bahdanau, Salem Lahlou, Lucas Willems, Chitwan Saharia, Thien Huu Nguyen, and Yoshua Bengio. 2019. BabyAI: A Platform to Study the Sample Efficiency of Grounded Language Learning. In 7th International Conference on Learning Representations . New Orleans, LA

    work page 2019

  6. [6]

    Wang-Zhou Dai, Qiu-Ling Xu, Yang Yu, and Zhi-Hua Zhou. 2019. Bridging Machine Learning and Logical Reasoning by Abductive Learning. In Advances in Neural Information Processing Systems 32 . Vancouver, Canada, 2811–2822

    work page 2019

  7. [7]

    Honghua Dong, Jiayuan Mao, Tian Lin, Chong Wang, Lihong Li, and Denny Zhou. 2019. Neural Logic Machines. In 7th International Conference on Learning KDD ’25, August 03–07, 2025, Toronto, ON Jie-Jing Shao, Hao-Ran Hao, Xiao-Wen Yang and Yu-Feng Li. Representations. New Orleans, LA

    work page 2019

  8. [8]

    Bin Fang, Shidong Jia, Di Guo, Muhua Xu, Shuhuan Wen, and Fuchun Sun. 2019. Survey of imitation learning for robotic manipulation. International Journal of Intelligent Robotics and Applications 3, 4 (2019), 362–369

    work page 2019

Show all 44 references
  1. [9]

    Richard Fikes and Nils J. Nilsson. 1971. STRIPS: A New Approach to the Application of Theorem Proving to Problem Solving. Artificial Intelligence 2, 3/4 (1971), 189–208

    1971

  2. [10]

    Maria Fox and Derek Long. 2003. PDDL2.1: An Extension to PDDL for Expressing Temporal Planning Domains. Journal of Artificial Intelligence Research 20 (2003), 61–124

    2003

  3. [11]

    Tanmay Gangwani, Joel Lehman, Qiang Liu, and Jian Peng. 2019. Learning Belief Representations for Imitation Learning in POMDPs. In Proceedings of the 35th Conference on Uncertainty in Artificial Intelligence , Vol. 115. Tel Aviv, Israel, 1061–1071

    2019

  4. [12]

    En-Hao Gao, Yu-Xuan Huang, Wen-Chao Hu, Xin-Hao Zhu, and Wang-Zhou Dai

  5. [13]

    Caelan Reed Garrett, Chris Paxton, Tomás Lozano-Pérez, Leslie Pack Kaelbling, and Dieter Fox. 2020. Online Replanning in Belief Space for Partially Observable Task and Motion Problems. In IEEE International Conference on Robotics and Automation. Paris, France, 5678–5684

    2020

  6. [14]

    Alfonso Emilio Gerevini. 2020. An Introduction to the Planning Domain Definition Language (PDDL): Book review. Artificial Intelligence 280 (2020), 103221

    2020

  7. [15]

    Tenenbaum, and Jiajun Wu

    Joy Hsu, Jiayuan Mao, Joshua B. Tenenbaum, and Jiajun Wu. 2023. What’s Left? Concept Grounding with Logic-Enhanced Foundation Models. In Advances in Neural Information Processing Systems 36 . New Orleans, LA

    2023

  8. [16]

    De-An Huang, Danfei Xu, Yuke Zhu, Animesh Garg, Silvio Savarese, Li Fei-Fei, and Juan Carlos Niebles. 2019. Continuous Relaxation of Symbolic Planner for One-Shot Imitation Learning. In 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems . Macau, China, 2635–2642

    2019

  9. [17]

    Muggleton, and Yuan Jiang

    Yu-Xuan Huang, Wang-Zhou Dai, Le-Wen Cai, Stephen H. Muggleton, and Yuan Jiang. 2021. Fast Abductive Learning by Similarity-based Consistency Optimization. In Advances in Neural Information Processing Systems 34 . Virtual Event, 26574–26584

    2021

  10. [18]

    Yu-Xuan Huang, Wang-Zhou Dai, Jian Yang, Le-Wen Cai, Shaofen Cheng, Ruizhang Huang, Yu-Feng Li, and Zhi-Hua Zhou. 2020. Semi-Supervised Abductive Learning and Its Application to Theft Judicial Sentencing. In 20th IEEE International Conference on Data Mining . Sorrento, Italy, ...

    2020

  11. [19]

    Yu-Xuan Huang, Zequn Sun, Guangyao Li, Xiaobin Tian, Wang-Zhou Dai, Wei Hu, Yuan Jiang, and Zhi-Hua Zhou. 2023. Enabling Abductive Learning to Exploit Knowledge Graph. In Proceedings of the 32nd International Joint Conference on Artificial Intelligence. Macao, China, 3839–3847

    2023

  12. [20]

    Ahmed Hussein, Mohamed Medhat Gaber, Eyad Elyan, and Chrisina Jayne. 2017. Imitation learning: A survey of learning methods. Comput. Surveys 50, 2 (2017)

    2017

  13. [21]

    Emily Jin, Jiaheng Hu, Zhuoyi Huang, Ruohan Zhang, Jiajun Wu, Li Fei- Fei, and Roberto Martín-Martín. 2023. Mini-BEHAVIOR: A Procedurally Generated Benchmark for Long-horizon Decision-Making in Embodied AI. CoRR abs/2310.01824 (2023)

    2023 arXiv

  14. [22]

    George Dimitri Konidaris, Leslie Pack Kaelbling, and Tomás Lozano-Pérez. 2018. From Skills to Symbols: Learning Symbolic Representations for Abstract High- Level Planning. Journal of Artificial Intelligence Research 61 (2018), 215–289

    2018

  15. [23]

    Matthews, Ivan Villa- Renteria, Jerry Huayang Tang, Claire Tang, Fei Xia, Silvio Savarese, Hyowon Gweon, C

    Chengshu Li, Ruohan Zhang, Josiah Wong, Cem Gokmen, Sanjana Srivastava, Roberto Martín-Martín, Chen Wang, Gabrael Levine, Michael Lingelbach, Jiankai Sun, Mona Anvari, Minjune Hwang, Manasi Sharma, Arman Aydin, Dhruva Bansal, Samuel Hunter, Kyu-Young Kim, Alan Lou, Caleb R. Ma...

    2022

  16. [24]

    Jiayuan Mao, Tomás Lozano-Pérez, Josh Tenenbaum, and Leslie Pack Kaelbling

  17. [25]

    Luc Le Mero, Dewei Yi, Mehrdad Dianati, and Alexandros Mouzakitis. 2022. A Survey on Imitation Learning Techniques for End-to-End Autonomous Vehicles. IEEE Transactions on Intelligent Transportation Systems 23, 9 (2022), 14128–14147

    2022

  18. [26]

    Yang, Jiantao Jiao, and Kannan Ramchandran

    Nived Rajaraman, Lin F. Yang, Jiantao Jiao, and Kannan Ramchandran. 2020. Toward the Fundamental Limits of Imitation Learning. In Advances in Neural Information Processing Systems 33 . Virtual Event

    2020

  19. [27]

    Jie-Jing Shao, Hao-Sen Shi, Lan-Zhe Guo, and Yu-Feng Li. 2024. Offline Imitation Learning with Model-based Reverse Augmentation. In Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining . Barcelona, Spain, 2608–2617

    2024

  20. [28]

    Jie-Jing Shao, Hao-Sen Shi, Tian Xu, Lan-Zhe Guo, Yang Yu, and Yu-Feng Li. 2024. Offline Imitation Learning without Auxiliary High-quality Behavior Data

    2024

  21. [29]

    Mohit Shridhar, Lucas Manuelli, and Dieter Fox. 2021. CLIPort: What and Where Pathways for Robotic Manipulation. In 5th Conference on Robot Learning. London, UK, 894–906

    2021

  22. [30]

    Tenenbaum, Tomás Lozano-Pérez, and Leslie Pack Kaelbling

    Tom Silver, Ashay Athalye, Joshua B. Tenenbaum, Tomás Lozano-Pérez, and Leslie Pack Kaelbling. 2022. Learning Neuro-Symbolic Skills for Bilevel Planning. In 6th Conference on Robot Learning . Auckland, New Zealand, 701–714

    2022

  23. [31]

    Tenenbaum

    Tom Silver, Rohan Chitnis, Nishanth Kumar, Willie McClinton, Tomás Lozano- Pérez, Leslie Pack Kaelbling, and Joshua B. Tenenbaum. 2023. Predicate Invention for Bilevel Planning. In 37th AAAI Conference on Artificial Intelligence . Washington, DC, 12120–12129

    2023

  24. [32]

    Tenenbaum, Leslie Pack Kaelbling, and Tomás Lozano-Pérez

    Tom Silver, Rohan Chitnis, Joshua B. Tenenbaum, Leslie Pack Kaelbling, and Tomás Lozano-Pérez. 2021. Learning Symbolic Operators for Task and Motion Planning. In IEEE/RSJ International Conference on Intelligent Robots and Systems . Prague, Czech Republic, 3182–3189

    2021

  25. [33]

    Chen Wang, Linxi Fan, Jiankai Sun, Ruohan Zhang, Li Fei-Fei, Danfei Xu, Yuke Zhu, and Anima Anandkumar. 2023. MimicPlay: Long-Horizon Imitation Learning by Watching Human Play. In 7th Annual Conference on Robot Learning . Atlanta, GA

    2023

  26. [34]

    Renhao Wang, Jiayuan Mao, Joy Hsu, Hang Zhao, Jiajun Wu, and Yang Gao

  27. [35]

    Grover J Whitehurst and Ross Vasta. 1975. Is language acquired through imitation? Journal of Psycholinguistic Research 4 (1975), 37–59

    1975

  28. [36]

    Danfei Xu, Roberto Martín-Martín, De-An Huang, Yuke Zhu, Silvio Savarese, and Li Fei-Fei. 2019. Regression Planning Networks. InAdvances in Neural Information Processing Systems 32. Vancouver, Canada, 1317–1327

    2019

  29. [37]

    Tian Xu, Ziniu Li, and Yang Yu. 2022. Error Bounds of Imitating Policies and Environments for Reinforcement Learning. IEEE Transactions on Pattern Analysis and Machine Intelligence 44, 10 (2022), 6968–6980

    2022

  30. [38]

    Xiaowen Yang, Jie-Jing Shao, Wei-Wei Tu, Yufeng Li, Wang-Zhou Dai, and Zhi- Hua Zhou. 2024. Safe Abductive Learning in the Presence of Inaccurate Rules. In 38th AAAI Conference on Artificial Intelligence. Vancouver, Canada, 16361–16369

    2024

  31. [39]

    Xiaowen Yang, Wenda Wei, Jie-Jing Shao, Yufeng Li, and Zhi-Hua Zhou. 2024. Analysis for Abductive Learning and Neural-Symbolic Reasoning Shortcuts. In 41st International Conference on Machine Learning . Vienna, Austria

    2024

  32. [40]

    Gordon, and Fei Fang

    Zhuorui Ye, Stephanie Milani, Geoffrey J. Gordon, and Fei Fang. 2024. Concept- Based Interpretable Reinforcement Learning with Limited to No Human Labels. CoRR abs/2407.15786 (2024)

    2024 arXiv

  33. [41]

    Zhi-Hua Zhou. 2019. Abductive learning: towards bridging machine learning and logical reasoning. SCIENCE CHINA Information Science 62, 7 (2019), 76101:1– 76101:3. A EXPERIMENTAL DETAILS A.1 BabyAI and Mini-BEHA VIOR In these two environments, each object feature is represented...

    2019

  34. [2022]

    In Advances in Neural Information Processing Systems 35

    PDSketch: Integrated Domain Programming, Learning, and Planning. In Advances in Neural Information Processing Systems 35 . New Orleans, LA

  35. [2023]

    In The 11th International Conference on Learning Representations

    Programmatically Grounded, Compositionally Generalizable Robotic Manipulation. In The 11th International Conference on Learning Representations . Kigali, Rwanda

  36. [2024]

    In Proceedings of the 38th AAAI Conference on Artificial Intelligence

    Knowledge-Enhanced Historical Document Segmentation and Recognition. In Proceedings of the 38th AAAI Conference on Artificial Intelligence

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.