An Thai Le

Research Interests

  • Scaling robot learning & planning
  • Dynamics generative models
  • Optimal Transport theory and gradient flows

Affiliation

TU Darmstadt, Intelligent Autonomous Systems, Computer Science Department

Contact

an.le@tu-darmstadt.de
Room D202, Building S2|02, TU Darmstadt, FB-Informatik, FG-IAS, Hochschulstr. 10, 64289 Darmstadt
+49-6151-16-20073

An Thai Le joined the Intelligent Autonomous System lab on November 1st, 2021, as a Ph.D. student. Currently, he is working to scale robotics learning and planning methods with batch optimization. In particular, he aims to scale planning methods to long-horizon, high-dimensional state-space, number of plans, and number of agents. Batch planning methods are crucial for robotics since they could discover many homotopy classes in the multi-objective problems, thereby exhibiting robustness to bad local minima. Furthermore, these batch methods can serve as a strong oracle for collecting datasets or striving to discover a global optimal solution for robotics skill execution. Besides his main works, he has applied entropic optimal transport in other ML applications, such as molecular representation aggregation for graph-based models or CLIP embedding alignments.

Reviewing

IROS, ICRA, R:SS, RLC, CoRL, NeurIPS, ICML, ICLR, AAAI, IEEE RA-L, IEEE T-RO, and various Robotics & ML workshops.

Teaching Assistant

  • Reinforcement Learning (SS 2022)
  • Statistical Machine Learning (SS 2023, WS 2023/2024, SS2024)
  • Robot Learning Integrated Project (WS 2024/2025)

Before his Ph.D., An Thai Le worked on forceful imitation learning during the internship at Bosch AI. He also worked on task and motion planning for human-robot collaboration settings. During his master's research, he was fortunate to be advised by Dr. Meng Guo and Dr. Jim Mainprice.

Recent Publications

Optimal Transport In Robotics

    •       Bib
      Le, A. T.; Chalvatzaki, G.; Biess, A.; Peters, J. (2023). Accelerating Motion Planning via Optimal Transport, Advances in Neural Information Processing Systems (NIPS / NeurIPS).
    •     Bib
      Le, A. T.; Chalvatzaki, G.; Biess, A.; Peters, J. (2023). Accelerating Motion Planning via Optimal Transport, IROS 2023 Workshop on Differentiable Probabilistic Robotics: Emerging Perspectives on Robot Learning, [Oral].
    •     Bib
      Le, A. T.; Chalvatzaki, G.; Biess, A.; Peters, J. (2023). Accelerating Motion Planning via Optimal Transport, NeurIPS 2023 Workshop Optimal Transport and Machine Learning, [Oral].
    •       Bib
      Le, A. T.; Hansel, K.; Peters, J.; Chalvatzaki, G. (2023). Hierarchical Policy Blending As Optimal Transport, 5th Annual Learning for Dynamics & Control Conference (L4DC), PMLR.

Optimal Transport In ML

    •     Bib
      Nguyen, D.M.H.; Lukashina, N.; Nguyen, N.; Le, A.T.; Nguyen, T.T.; Ho, N.; Peters, J.; Sonntag, D.; Zaverkin, V.; Niepert, M. (2024). Structure-Aware E(3)-Invariant Molecular Conformer Aggregation Networks, Proceedings of the International Conference on Machine Learning (ICML).
    •     Bib
      Nguyen, D.H.M.*; Le, A.T.*; Nguyen, T.Q.; Nghiem, T.D.; Duong-Tran, D. ; Peters, J.; Li, S.; Niepert, M.; Sonntag, D. (2024). Dude: Dual Distribution-Aware Context Prompt Learning For Large Vision-Language Model, Asian Conference on Machine Learning (ACML).

Generative Modeling For Imitation Learning And Motion Planning

    •       Bib
      Carvalho, J.; Le, A. T.; Baierl, M.; Koert, D.; Peters, J. (2023). Motion Planning Diffusion: Learning and Planning of Robot Motions with Diffusion Models, IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).
    •       Bib
      Urain, J.*; Le, A.T.*; Lambert, A.*; Chalvatzaki, G.; Boots, B.; Peters, J. (2022). Learning Implicit Priors for Motion Optimization, IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).
    •   Bib
      Le, A. T.; Urain, J.; Lambert, A.; Chalvatzaki, G.; Boots, B.; Peters, J. (2022). Learning Implicit Priors for Motion Optimization, RSS 2022 Workshop on Implicit Representations for Robotic Manipulation.
    •       Bib
      Le, A. T.; Guo M.; Duijkeren, N.; Rozo, L.; Krug, R.; Kupcsik, A.G.; Buerger, M. (2021). Learning forceful manipulation skills from multi-modal human demonstrations, IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

Vectorization For Planning

    •   Bib
      Le, A. T.; Hansel, K.; Carvalho, J.; Urain, J.; Biess, A.; Chalvatzaki, G.; Peters, J. (2024). Global Tensor Motion Planning, CoRL 2024 Workshop on Differentiable Optimization Everywhere.

Supervised Theses and Projects

YearTypeTogether withStudent(s)TopicDocument
2022ProjectJunning HuangLu Liu, Yuheng Ouyang, Jiahui ShiBenchmark Neural Lyapunov Control Algorithms on Pendulum
2022ProjectJunning HuangChao Jin, Liyuan Xiang, Peng YanHybrid Motion-Force Optimization
2022Master ThesisAli Younes, Georgia ChalvatzakiDaljeet NandhaLeveraging Large Language Models For Autonomous Task Planningpdf
2023Master ThesisJoao Carvalho, Julen Urain De JesusMark BaierlScore-based Planning Networks for Robot Motion Planningpdf
2023Master ThesisKay Hansel, Jan Peters, Georgia ChalvatzakiAlper GeceLeveraging Structured-Graph Correspondence in Imitation Learning
2023Master ThesisKay Hansel, Jan PetersMarius ZoellerEnhancing Smoothness in Policy Blending with Gaussian Processes[Ongoing]
2023Master ThesisAli Younes, Georgia ChalvatzakiCaio FreitasGraph Correspondence Diffusion For Imitation Learning
2023Master ThesisGeorgia ChalvatzakiDenis AndricLearning Symplectic Manifold Of Dynamical Systems[Ongoing]
2023Master ThesisGeorgia ChalvatzakiNico NonnengiesserGraph Neural Network For Robotics Control[Ongoing]
2024Master ThesisJoao CarvalhoQiao SunGeometry-Aware Diffusion Models for Robotics[Ongoing]
2024Master ThesisKay HanselSebastian ZachReactive Motion Generation through Probabilistic Dynamic Graphs[Ongoing]