Ashley Edwards
Ashley Edwards is an AI researcher who studies how agents can learn from video and helps turn videos into interactive worlds. She earned a computer science degree at the University of Georgia and a PhD at Georgia Tech, then worked at Uber AI Labs and Google DeepMind. She coauthored Gato and worked on Genie, which won an ICML 2024 Best Paper award. She worked on video generation at Runway in 2025 and returned to Google DeepMind by February 2026. As of October 2026, she is a senior research scientist focused on reinforcement learning and foundational world models.
- Now at
- Google DeepMind
- Badges
Timeline
-
2011–2019
A Georgia Tech PhD
She earns a BS in computer science at the University of Georgia in 2011 and completes a PhD at Georgia Tech in 2019. Her doctoral work develops models that infer latent actions, rewards, and policies from videos.
-
2018–2019
Learning policies from video
In 2018, she interns at Google Brain. In 2019, she coauthors the ICML paper Imitating Latent Policies from Observation, which infers latent policies directly from state observations.
-
March 2019
Research at Uber AI Labs
She joins Uber AI Labs as a research scientist in March 2019.
-
2020
Value from state transitions
Her ICML 2020 paper, Estimating Q(s,s’) with Deep Deterministic Dynamics Gradients, defines value over neighboring state transitions and learns a forward dynamics model to maximize it.
-
2022
A generalist agent
She coauthors A Generalist Agent in 2022, introducing Gato, a multi-modal, multi-task, multi-embodiment policy that can play Atari, caption images, chat, and control a robot arm with the same network.
-
2024
Genie wins an ICML award
By 2024, she is a senior research scientist at Google DeepMind and works on Genie, an unsupervised generative model trained on unlabelled Internet videos to create action-controllable worlds. Genie wins an ICML Best Paper award in 2024.
-
March 2025
Video generation at Runway
In March 2025, she is a member of technical staff at Runway and works on video generation. Her research explores models that infer latent actions, rewards, and policies from videos alone.
-
As of October 2026
Back at Google DeepMind
By February 2026, she is back at Google DeepMind as a senior research scientist focused on reinforcement learning and foundational world models.
Career
- Since 2026Google DeepMindSenior Research Scientist
- From 2025RunwayMember of Technical Staff
- From 2019UberResearch Scientist
- From 2018Google BrainIntern
Education
- 2019Georgia Institute of TechnologyPhD
- 2011University of GeorgiaBS in computer science
Sources and credits
Photo: We have no published photo of Ashley Edwards yet. Send us a free-licensed photo.
- Ashley Edwards
- News — Ashley Edwards
- Imitating Latent Policies from Observation | ML Anthology
- Estimating Q(s,s’) with Deep Deterministic Dynamics Gradients
- A Generalist Agent
- Announcing Ashley Edwards at RAAIS 2024
- ICML 2024 Awards
- ICARL - Seminar Series
- Ashley Edwards – RPL Summer School 2026
- Speaker Series Winter 2025/26: eight talks to rewatch





