Fwd: Learning in an Agentic World - Workshop at COLT 2026

[Shaddin Dughmi]

See below for an interesting workshop being held in conjunction with COLT in San Diego next month.

---------- Forwarded message ----------

===============================================

We are excited to announce the upcoming workshop “Learning in an Agentic World” to be held at COLT 2026 (June 29, San Diego), which will bring together researchers working on the foundations of learning, AI agents, reasoning, reliability, adaptivity, strategic interaction, and generative AI systems.

The workshop aims to foster discussion across machine learning, learning theory, optimization, algorithms, economics/game theory, and AI safety, with topics including:

• Learning and decision-making with AI agents
• Reliability, verification, and evaluation of LLMs
• Strategic and adaptive learning environments
• Personalization and human-AI interaction
• Theoretical foundations for reasoning and generation

The workshop will feature invited talks, contributed poster presentations, and opportunities for broad cross-community interaction. Consider submitting your work to our call for abstracts (deadline June 1).

More details, including speakers, schedule, and submission information, are available on the workshop website.

We look forward to your participation.

Organizers:

Hedyeh Beyhaghi (UMass Amherst), Avrim Blum (TTIC), Han Shao (University of Maryland), Dravy Sharma (TTIC)

Shaddin Dughmi

Associate Professor

Computer Science Department

University of Southern California (USC)
Email: sha...@usc.edu

Web: https://viterbi-web.usc.edu/~shaddin/