Responsive image

Siddharth Srivastava

Associate Professor

School of Computing and Augmented Intelligence

Arizona State University

siddharths/at/asu/dot/edu

#592, 699 S Mill Ave.
Tempe, AZ 85282



Research areas:     Generalizable planning and learning in AI and Robotics, AI assessment

My research focuses on principles and algorithms for reliable AI systems that can carry out generalizable planning and learning to solve complex, user-desired tasks. My group is currently working on topics such as learning abstraction hierachies, transfer learning, learning representations for sequential decision-making and on new paradigms for third-party safety and capability assessment of AI systems that can learn and plan. We are particularly interested in developing principles for AI systems that can reliably (and safely) learn, generalize and transfer knowledge to solve new, user-desired tasks not encountered during training.


Check out my group's webpage to learn more about our research!


Student opportunities: If you are interested in joining my group, please send me an email with your CV, a copy of your transcript, and a note on how you might contribute to some of our projects. If you have done well in technically rigorous classes in AI, or mathematics, or CS theory, you are already off to a great start!

Press Coverage: The BBC | Motherboard | AAAS Eureka Alert! Science News | NBC News | IEEE Spectrum

Recent Events and Highlights

We are organizing the first symposium focused on Assessment of adaptive AI systems as a part of the AAAI Spring Symposia Series at Stanford, March 25-27 2024.

Join us and participate in formative discussions on the topic!

1/24: I was honored to receive the Top 5% Faculty Award for my graduate-level class on Artificial Intelligence.

Thanks to all the participants and co-organizers who helped make the seventh edition of the GenPlan Workshop at NeurIPS a great success. We had 200+ attendees with an enriching diversity of research approaches on the topic.

11/23: It was a pleasure talking to Sean Leahy about AI Safety and Assessment at the Learning Futures podcast! Check out the episode here or here.

9/23: PhD students Pulkit Verma and Rushang Karia will be presenting our research on capability assessment of AI systems in stochastic settings at NeurIPS in December.

8/23: PhD student Naman Shah and I presented a tutorial on Integrated Task and Motion Planning at IJCAI 2023. [Slides]

Our work with Michael James McDonald, Mitch Lin, Dylan Hadfield-Menell, Pieter Abbeel and Stuart Russell was featured by the BBC (pdf).

I served as conference chair of ICAPS 2019.