Theoretical Aspects of Trustworthy AI

Location

Calvin Lab auditorium

Date

Monday, Apr. 28 – Friday, May 2, 2025

Back to calendar

All talks listed in Pacific Time. Schedule subject to change.

8:30 – 9 a.m.

Coffee and Check-In

9 – 9:45 a.m.

Introduction

Sanjit Seshia (UC Berkeley)

9:45 – 10:30 a.m.

Veridical Data Science towards Trustworthy AI

Bin Yu (UC Berkeley)

10:30 – 11 a.m.

Break

11 – 11:45 a.m.

Veridical Data Science towards Trustworthy AI

Bin Yu (UC Berkeley)

12 – 1:30 p.m.

Lunch (on your own)

1:30 – 2:30 p.m.

Provably Safe and Beneficial AI

Stuart Russell (UC Berkeley)

2:30 – 3 p.m.

Safety of GenAI through the lens of Security and Cryptography

Somesh Jha (University of Wisconsin-Madison)

3 – 3:30 p.m.

Break

3:30 – 4:30 p.m.

Safety of GenAI through the Lens of Security and Cryptography

Somesh Jha (University of Wisconsin-Madison)

4:30 – 5:30 p.m.

Reception

8:30 – 9 a.m.

Coffee and Check-In

9 – 9:45 a.m.

Query Languages for Machine Learning Models

Pablo Barcelo (Pontificia Universidad Catolica de Chile)

9:45 – 10:30 a.m.

Symbolic Reasoning about Large Language Models

Guy van den Broeck (UCLA)

10:30 – 11 a.m.

Break

11 – 11:45 a.m.

Learning and decision making under observer effects

Sarah Dean (Cornell University)

11:45 a.m. – 2 p.m.

Lunch (on your own)

2 – 2:45 p.m.

Towards Reasoning with a Million Environment Models

Sebastian Junges (Radboud University)

2:45 – 3:15 p.m.

Break

3:15 – 4 p.m.

Automata Embeddings for Goal-Conditioned Reinforcement Learning

Beyazit Yalcinkaya (UC Berkeley)

8:30 – 9 a.m.

Coffee and Check-In

9 – 9:45 a.m.

Machine Teaching: Supervised Learning and Beyond

Jerry Zhu (University of Wisconsin-Madison)

9:45 – 10:30 a.m.

Language Model Guided Synthesis for Lifting

Elizabeth Polgreen (University of Edinburgh)

10:30 – 11 a.m.

Break

11 – 11:45 a.m.

Neurosymbolic Synthesis for Trustworthy Machine Learning

Osbert Bastani (University of Pennsylvania)

11:45 a.m. – 2 p.m.

Lunch (on your own)

2 – 2:45 p.m.

Formal Models of Machine Teaching Without Collusion (Remote Talk)

Sandra Zilles (University of Regina)

2:45 – 3:15 p.m.

Break

3:15 – 4 p.m.

Trustworthy AI for Legal Reasoning

Ruzica Piskac (Yale University)

8:30 – 9 a.m.

Coffee and Check-In

9 – 9:45 a.m.

Robust Learning with Principled Out-of-Distribution Detection via Multiple Testing

Susmit Jha (SRI International)

9:45 – 10:30 a.m.

Exploiting Epistemic Uncertainty for Deep Exploration in Reinforcement Learning

Matthijs Spaan (Delft University of Technology)

10:30 – 11 a.m.

Break

11 – 11:45 a.m.

Prediction-Powered Inference

Anastasios Angelopoulos

12 – 2 p.m.

Lunch (on your own)

2 – 2:45 p.m.

Leveraging Structure in Formal Logic-Guided Reinforcement Learning

Ameesh Shah (UC Berkeley)

2:45 – 3:15 p.m.

Break

3:15 – 4 p.m.

Shielded Reinforcement Learning

Bettina Koenighofer (Technical University of Graz)

8:30 – 9 a.m.

Coffee and Check-In

9 – 9:45 a.m.

AI for Safety Critical Control

Claire Tomlin (UC Berkeley)

9:45 – 10:30 a.m.

Edit Distance & Robustness in (In)finite Word Languages

Dana Fisman (Ben-Gurion University)

10:30 – 11 a.m.

Break

11 – 11:45 a.m.

Learning and Teaching Automata via Demonstrations and Natural Languages

Marcell Vazquez-Chanlatte (Nissan Advanced Technology Center - Silicon Valley)

11:45 a.m. – 2 p.m.

Lunch (on your own)

2 – 2:45 p.m.

Synthesizing Pareto-optimal Interpretations of Black Box ML Models

Supratik Chakraborty (IIT Bombay)

2:45 – 3:15 p.m.

Break

3:15 – 4 p.m.

Robust Algorithmic Recourse with Predictions

Shahin Jabbari (Drexel University)