I am a computer scientist (PhD) researching Artificial Intelligence (AI) and Machine Learning (ML). Currently, I work as Senior Computer Scientist at SRI International, in the Computer Vision lab in its Princeton, NJ, USA office.
I am a computer scientist (PhD) researching Artificial Intelligence (AI) and Machine Learning (ML). Currently, I work as Senior Computer Scientist at SRI International, in the Computer Vision lab in its Princeton, NJ, USA office.
Research Interests
Efficient AI/ML: I believe in pushing the boundaries of energy-efficient AI/ML. Reducing the energy consumption for the training and inference of AI models is paramount to sustainable and scalable usage of current and future frontier models. My current research includes analog neural networks (funded by DARPA), photonic neural networks (funded by NRO), and biological neural networks (looking for partners). These embody the philosophy of 'using nature as a computer' , but the challenges of scalability and reliability need to be overcome. In the past, I also worked on digital edge computing such as learning in quantized NNs and Hyper-Dimensional Computing (HDC).
Reinforcement Learning (RL) and Generative Models: Advancing fundamental research in RL and multi-agent RL can unlock real-world and embodied autonomy. I work on continual and lifelong RL (funded by DARPA), and robust multi-agent RL (funded by Army Research Lab) using generative models, model-free lifelong RL and learning inter-agent communication for distributed state estimation and coordinated collective problem solving. Recently, I am interested in applying such methods to Large Language Model (LLM) fine-tuning and multi-agent AI systems (or agentic AI).
Mucllari, Edison, Aswin Raghavan, and Zachary Alan Daniels. "Class-incremental SAR ATR in noisy and adversarial environments." In Automatic Target Recognition XXXV, vol. 13463, p. 1346302. SPIE, 2025. (Paper)
Sur, Indranil, Aswin Raghavan, Abrar Rahman, James Z. Hare, Daniel Cassenti, and Carl Busart. "Data-Driven Distributed Common Operational Picture from Heterogeneous Platforms using Multi-Agent Reinforcement Learning." International Command and Control Research and Technology Symposium (ICCRTS) 2024 (Paper).
Farkya, Saurabh, Zachary Alan Daniels, Aswin Raghavan, Gooitzen van der Wal, Michael Isnardi, Michael Piacentino, and David Zhang. "Data-Driven Pixel Control: Challenges and Prospects." Dynamic Data Driven Applications Systems (DDAS 2024) (Paper).
Jun Hu, Phil Miller, Michael Lomnitz, Saurabh Farkya, Emre Yilmaz, Aswin Raghavan, David Zhang, Michael Piacentino. "Tools Identification By On-Board Adaptation of Vision-and-Language Models". To appear in the AAAI-24 Demonstrations Program at the Thirty-Eighth AAAI Conference on Artificial Intelligence (AAAI-24).
Farkya, Saurabh, Aswin Raghavan, and Avi Ziskind. "Improving the Robustness of Quantized Deep Neural Networks to White-Box Attacks using Stochastic Quantization and Information-Theoretic Ensemble Training" arXiv preprint arXiv:2312.00105 (2023). (arxiv)
Peihong Yu, Bhoram Lee, Aswin Raghavan, Supun Samarasekera, Pratap Tokekar (U. of Maryland), James Zachary Hare (DEVCOM Army Research Lab), "Enhancing Multi-Agent Coordination through Common Operating Picture Integration", Out-of-Distribution Generalization in Robotics (OOD Workshop) at Conference on Robot Learning (CoRL) 2023. (Paper)
Indranil Sur, Zachary A Daniels, Aswin Raghavan, Jesse Hostetler, Abrar Rahman, Michael R Piacentino, Ajay Divakaran, Roberto Corizzo (American University), Kamil Faber (AGH University of Science and Technology), Nathalie Japkowicz (American University), Michael Baron (American University), James S Smith (Georgia Institute of Technology), Sahana Joshi (Georgia Institute of Technology), Zsolt Kira (Georgia Institute of Technology), Tyler L Hayes (RIT), Christopher Kanan (University of Rochester), Gianmarco J Gallardo (RIT). "System Design for an Integrated Lifelong Reinforcement Learning Agent For Real-Time Strategy Games". Accepted for oral presentation at the Second International Conference on AI-ML Systems (2022). (conf)
Daniels, Zachary, Aswin Raghavan, Jesse Hostetler, Abrar Rahman, Indranil Sur, Michael Piacentino, and Ajay Divakaran. "Model-Free Generative Replay for Lifelong Reinforcement Learning: Application to Starcraft-2." In Proceedings of The First Conference on Lifelong Learning Agents (CoLLAs) (2022) PMLR (Paper)
Farkya, Saurabh; Daniels, Zachary A; Raghavan, Aswin; Zhang, David Zhang; Piacentino, Michael. Saccade Mechanisms for Image Classification, Object Detection and Tracking. NeuroVision workshop at CVPR 2022. (Page) (PDF)
Indhumathi Kandaswamy, Saurabh Farkya, Zachary Daniels, Gooitzen van der Wal, Aswin Raghavan, Yuzheng Zhang, Jun Hu, Michael Lomnitz, Michael Isnardi, David Zhang, Michael Piacentino. Real-Time Hyper-Dimensional Reconfiguration at the Edge using Hardware Accelerators, Embedded Vision Workshop (EVW) at CVPR 2022. (link)
Raghavan, A., Hostetler, J., Sur, I., Rahman, A., & Divakaran, A. Lifelong Learning using Eigentasks: Task Separation, Skill Acquisition, and Selective Transfer. Lifelong Machine Learning Workshop, International Conference on Machine Learning (ICML 2020). (workshop) (paper) (video)
He, Zecheng, Aswin Raghavan, Sek Chai, and Ruby Lee Power-Grid Controller Anomaly Detection with Enhanced Temporal Deep Learning TrustCom 2019 (18th IEEE International Conference on Trust, Security and Privacy in Computing and Communications) (arxiv)
Meo, Timothy J., Chris Kim, Aswin Raghavan, Alex Tozzo, David A. Salter, Amir Tamrakar, and Mohamed R. Amer. "Aesop: A visual storytelling platform for conversational AI and common sense grounding." AI Communications Preprint (2019): 1-18. (PDF)
Samyak Parajuli and Aswin Raghavan and Sek Chai, Generalized Ternary Connect: End-to-End Learning and Compression of Multiplication-Free Deep Neural Networks, (arxiv)
Durga Harish Dayapule, Aswin Raghavan, Prasad Tadepalli, Alan Fern, Emergency Response Optimization using Online Hybrid Planning, 27th International Joint Conference on Artificial Intelligence and the 23rd European Conference on Artificial Intelligence (IJCAI-ECAI 2018) (PDF)
Tharindu Mathew, Aswin Raghavan, Sek Chai, Event Prediction in Processors using Deep Temporal Models, 1st Workshop on Energy Efficient Machine Learning And Cognitive Computing for Embedded Applications (EMC2), 23rd ACM Intl. Conf. on Architectural Support for Programming Languages and Operating Systems (ASPLOS 2018) (IEEE)
Aswin Raghavan, Scott Sanner, Roni Khardon, Prasad Tadepalli, Alan Fern, Hindsight Optimization for Hybrid State and Action MDPs, Proceedings of the 31st AAAI Conference on Artificial Intelligence (AAAI-2017). (PDF)
Aswin Raghavan, Mohamed R. Amer, Timothy Shields, David Zhang, Sek Chai (SRI International), GPU Activity Prediction using Representation Learning, ML Systems Workshop, International Conference on Machine Learning (ICML), 2016. (PDF)
Sek Chai, Aswin Raghavan, David Zhang, Mohamed Amer, Tim Shields, Low Precision Neural Networks using Subband Decomposition, Presented at CogArch Workshop, Atlanta, GA, April 2016. (PDF)
Aswin Raghavan, Prasad Tadepalli, Alan Fern, Roni Khardon, Memory-Efficient Symbolic Online Planning for Factored MDPs, 31st Conference on Uncertainty in Artificial Intelligence (UAI), 2015. (PDF) (Poster)
A. Raghavan, A. Fern, P. Tadepalli, and R. Khardon, Symbolic Opportunistic Policy Iteration for Factored-Action MDPs, Proceedings of the International Conference on Neural Information Processing Systems (NIPS), 2013. (PDF)(BibTex)(Poster)
S. Joshi, R. Khardon, P. Tadepalli, A. Fern, A. Raghavan, Relational Markov Decision Processes: Promise and Prospects, StarAI Workshop help at the Twenty-Seventh AAAI National Conference on Artificial Intelligence (StarAI), 2013 (PDF).
S. Joshi, R. Khardon, P. Tadepalli, A. Raghavan, A. Fern, Solving Relational MDPs with Exogenous Events and Additive Rewards, The European Conference on Machine Learning (ECML/PKDD) , 2013 (PDF) (Arxiv).
Aswin Raghavan, Saket Joshi, Alan Fern, Prasad Tadepalli, Roni Khardon, Planning in Factored action spaces using Symbolic Dynamic Programming, Proceedings of the 26th Conference on Artificial Intelligence (AAAI-12) Toronto, Canada. (source)(pdf)(poster).
Aswin N. R., Manimaran S. S., Harini A., and Ravindran, B. (2010), Accurate Mobile Robot Localization in Indoor environments using Bluetooth. In the Proceedings of the 2010 IEEE International Conference on Robotics and Automation (ICRA 2010), pp. 4391-4396. IEEE Press. (PDF).
R.Malmathanraj, Aswin N Raghavan, V.Srivas and R.Gowtham Rangarajan, Mammogram tumor classification using Q learning based thresholding, BEATS 2010 : International Conference on Biomedical Engineering and Assistive Technologies, NIT JALANDHAR.
Patents
Amer, Mohamed R., Timothy J. Meo, Aswin Nadamuni Raghavan, et al. Artificial intelligence in interactive storytelling. 2020.
Chai, Sek M., Zecheng He, Aswin Nadamuni Raghavan, and Ruby B. Lee. Anomalous behavior detection in processor based systems. 2022.
Chai, Sek M., David C. Zhang, Mohamed R. Amer, Timothy J. Shields, Aswin Nadamuni Raghavan, and Bhaskar Ramamurthy. Systems and methods for optimizing operations of computing devices using deep neural networks. 2022.
Chai, Sek Meng, Aswin Nadamuni Raghavan, and Samyak Parajuli. Dynamic adaptation of deep neural networks. 2022.
Chai, Sek Meng, David Zhang, Mohamed Amer, Timothy J. Shields, and Aswin Nadamuni Raghavan. Low precision neural networks using subband decomposition. 2023.
Daniels, Zachary A., Jun Hu, Michael R. Lomnitz, et al. Adaptable and continually learning neural network architecture. 2025.
Das, Subhodev, Aswin Nadamuni Raghavan, Avraham Joshua Ziskind, et al. Generative Artificial intelligence for Explainable Collaborative and Competitive Problem Solving. 2023.
Raghavan, Aswin Nadamuni, Saurabh Farkya, Jesse Albert Hostetler, et al. Hardening a deep neural network against adversarial attacks using a stochastic ensemble. 2024.
Raghavan, Aswin Nadamuni, Jesse Hostetler, Indranil Sur, Abrar Abdullah Rahman, and Sek Meng Chai. Generative memory for lifelong machine learning. 2022.
Raghavan, Aswin Nadamuni, Jun Hu, David C. Zhang, et al. Object detection and visual grounding at the edge. 2025.
Raghavan, Aswin Nadamuni, Michael R. Piacentino, Michael A. Isnardi, et al. Reconfigurable, hyperdimensional neural network architecture. 2024.
Raghavan, Aswin Nadamuni, Indranil Sur, Zachary Daniels, et al. System design for an integrated lifelong machine learning agent. 2024.
Raghavan, Aswin Nadamuni, David Chao Zhang, Saurabh Farkya, et al. Energy efficient machine learning on the edge with query-based knowledge assistance. 2025.
Velipasalar, Senem, Sek Meng Chai, and Aswin Nadamuni Raghavan. Low power and privacy preserving sensor platform for occupancy detection. 2023.
Zhang, David Chao, Michael R. Piacentino, and Aswin Nadamuni Raghavan. Edge device having a heterogenous neuromorphic computing architecture. 2022.
Zhang, David Chao, Michael R. Piacentino, and Aswin Nadamuni Raghavan. Video processor capable of in-pixel processing. 2023.
Previous Experience