Anaïs Berkes

PhD student, Computer Science and Technology, University of Cambridge

Joined

May 2025

Names

Anaïs Berkes (Preferred)

Emails

****@cam.ac.uk (Confirmed)

Personal Links

Homepage

Google Scholar

ORCID

Career & Education History

PhD student

Computer Science and Technology , University of Cambridge (cam.ac.uk)

2023 – 2026

Advisors, Relations & Conflicts

PhD Advisor

Robert D. Mullins

Present

Coauthor

David Rolnick

Present

Coauthor

Yoshua Bengio

Present

Coauthor

Donna Vakalis

Present

Coauthor

Vincent Taboga

Present

Expertise

AI Safety

2024 – 2025

Meta Learning

2023 – 2025

Reinforcement Learning

2019 – 2025

Publications

In-Context Reinforcement Learning through Bayesian Fusion of Context and Value Prior
Anaïs Berkes, Vincent Taboga, Donna Vakalis, David Rolnick, Yoshua Bengio
- Submitted to ICLR 2026
- Readers: Everyone
HVAC-SPICE: Value-Uncertainty In-Context RL with Thompson Sampling for Zero-Shot HVAC Control
Anaïs Berkes, Vincent Taboga, Donna Vakalis, David Rolnick, Yoshua Bengio
- urbanai Poster
- Readers: Everyone
Graph Dreamer: Temporal Graph World Models for Sample-Efficient and Generalisable Reinforcement Learning
Anaïs Berkes, Donna Vakalis, Yoshua Bengio, David Rolnick
- WiML @ NeurIPS 2025
- Readers: Everyone
HVAC-GRACE: Transferable Building Control via Heterogeneous Graph Neural Network Policies
Anaïs Berkes, Donna Vakalis, David Rolnick, Yoshua Bengio
- CO-BUILD Poster
- Readers: Everyone

Anaïs Berkes

Names

Emails

Personal Links

Career & Education History

Advisors, Relations & Conflicts

Expertise

Publications

In-Context Reinforcement Learning through Bayesian Fusion of Context and Value Prior

HVAC-SPICE: Value-Uncertainty In-Context RL with Thompson Sampling for Zero-Shot HVAC Control

Graph Dreamer: Temporal Graph World Models for Sample-Efficient and Generalisable Reinforcement Learning

HVAC-GRACE: Transferable Building Control via Heterogeneous Graph Neural Network Policies

Co-Authors