Publications

under review

Critical Option-Critic via Distributional RLSubrat Prasad Panda, Mohit Prashant, and Arvind Easwaran RLHierarchical RL

@unpublished{panda2027b,
  title   = {Critical Option-Critic via Distributional {RL}},
  author  = {Panda, Subrat Prasad and Prashant, Mohit and Easwaran, Arvind},
    year    = {2027}
}

Learning Abstractions for the Safety Verification of RL PoliciesSubrat Prasad Panda, Liao Yuke, Blaise Genest, Djordje Zikelic, and Arvind Easwaran VerificationNeurosymbolicRL

@unpublished{panda2027a,
  title   = {Learning Abstractions for the Safety Verification of RL Policies},
  author  = {Panda, Subrat Prasad and Yuke, Liao and Genest, Blaise and Zikelic, Djordje and Easwaran, Arvind},
    year    = {2027}
}

Neurosymbolic Learning of Interpretable Abstraction in Reinforcement LearningSubrat Prasad Panda, Blaise Genest, and Arvind Easwaran NeurosymbolicRL

@unpublished{panda2027_workshop_a,
  title   = {Neurosymbolic Learning of Interpretable Abstraction in Reinforcement Learning},
  author  = {Panda, Subrat Prasad and Genest, Blaise and Easwaran, Arvind},
    year    = {2027}
}

published

2026 · 2 papers

Approximation-Free Differentiable Oblique Decision TreesSubrat Prasad Panda, Blaise Genest, and Arvind Easwaran Journal of Machine Learning Research (JMLR) journal NeurosymbolicRL paper code

@article{panda2025jmlr,
  url     = {http://jmlr.org/papers/v27/25-2047.html},
  title   = {Approximation-Free Differentiable Oblique Decision Trees},
  author  = {Panda, Subrat Prasad and Genest, Blaise and Easwaran, Arvind},
  journal = {Journal of Machine Learning Research (JMLR)},
  volume  = {27},
  year    = {2026}
}

Neurosymbolic Reasoning with Incremental Knowledge for Sample Efficient Hierarchical Reinforcement LearningSubrat Prasad Panda, Blaise Genest, and Arvind Easwaran ECML-PKDD conference RLHierarchical RLNeurosymbolicPlanningKnowledge arxiv code

@inproceedings{panda2026a,
  title   = {Neurosymbolic Reasoning with Incremental Knowledge for Sample Efficient Hierarchical Reinforcement Learning},
  author  = {Panda, Subrat Prasad and Genest, Blaise and Easwaran, Arvind},
  booktitle = {ECML-PKDD},
  year    = {2026}
}

2025 · 1 paper

Programmatic Reinforcement Learning for Trustworthy Microgrid ManagementSubrat Prasad Panda, Blaise Genest, and Arvind Easwaran AI4X International Conference workshop RLNeurosymbolic paper

@misc{panda2025programmatic,
  url          = {https://openreview.net/forum?id=S27okPWTtk},
  title        = {Programmatic Reinforcement Learning for Trustworthy Microgrid Management},
  author       = {Panda, Subrat Prasad and Genest, Blaise and Easwaran, Arvind},
  year         = {2025},
  howpublished = {AI4X International Conference},
  note         = {}
}

2024 · 2 papers

Methods for Mitigating Uncertainty in Real-Time Operations of a Connected MicrogridSubrat Prasad Panda, Blaise Genest, Arvind Easwaran, Rémy Rigo-Mariani, and Pengfeng Lin Sustainable Energy, Grids and Networks journal RL paper

@article{panda2024methods,
  doi={10.1016/j.segan.2024.101334},
  title={Methods for Mitigating Uncertainty in Real-Time Operations of a Connected Microgrid},
  author={Panda, Subrat Prasad and Genest, Blaise and Easwaran, Arvind and Rigo-Mariani, R{\'e}my and Lin, Pengfeng},
  journal={Sustainable Energy, Grids and Networks},
  volume={38},
  pages={101334},
  year={2024},
  publisher={Elsevier}
}

Vanilla Gradient Descent for Oblique Decision TreesSubrat Prasad Panda, Blaise Genest, Arvind Easwaran, and Ponnuthurai Nagaratnam Suganthan European Conference on Artificial Intelligence (ECAI) conference NeurosymbolicRL arxiv paper code project

@inproceedings{dtsemnet,
  title = {Vanilla Gradient Descent for Oblique Decision Trees},
  url = {http://dx.doi.org/10.3233/FAIA240607},
  DOI = {10.3233/faia240607},
  booktitle = {European Conference on Artificial Intelligence (ECAI)},
  publisher = {IOS Press},
  author = {Panda,  Subrat Prasad and Genest,  Blaise and Easwaran,  Arvind and Suganthan,  Ponnuthurai Nagaratnam},
  year = {2024},
  month = oct 
}

2021 · 2 papers

User Allocation in Mobile Edge Computing: A Deep Reinforcement Learning ApproachSubrat Prasad Panda, Ansuman Banerjee, and Arani Bhattacharya 2021 IEEE International Conference on Web Services (ICWS) conference RLEdge AI code

@inproceedings{panda2021user,
  title={User Allocation in Mobile Edge Computing: A Deep Reinforcement Learning Approach},
  author={Panda, Subrat Prasad and Banerjee, Ansuman and Bhattacharya, Arani},
  booktitle={2021 IEEE International Conference on Web Services (ICWS)},
  pages={447--458},
  year={2021},
  organization={IEEE}
}

Service Allocation/Placement in Multi-Access Edge Computing with Workload FluctuationsSubrat Prasad Panda, Kaustabha Ray, and Ansuman Banerjee International Conference on Service-Oriented Computing conference Edge AI

@inproceedings{panda2021service,
  title={Service Allocation/Placement in Multi-Access Edge Computing with Workload Fluctuations},
  author={Panda, Subrat Prasad and Ray, Kaustabha and Banerjee, Ansuman},
  booktitle={International Conference on Service-Oriented Computing},
  pages={747--755},
  year={2021},
  organization={Springer}
}

2020 · 1 paper

Dynamic Edge User Allocation with User Specified QoS PreferencesSubrat Prasad Panda, Kaustabha Ray, and Ansuman Banerjee International Conference on Service-Oriented Computing conference Edge AI

@inproceedings{panda2020dynamic,
  title={Dynamic Edge User Allocation with User Specified {QoS} Preferences},
  author={Panda, Subrat Prasad and Ray, Kaustabha and Banerjee, Ansuman},
  booktitle={International Conference on Service-Oriented Computing},
  pages={187--197},
  year={2020},
  organization={Springer}
}