RL as a Model of Agency Workshop at RLDM 2022

28 Jul 2022

Brian Christian gave a talk titled “Normalization and Value” at the RL as a Model of Agency Workshop.

Simons Institute, AI and Humanity Workshop at UC Berkeley 

16 Jul 2022

As a part of the Simons Institute’s Summer Cluster, a workshop, AI and Humanity, took place at UC Berkeley from July 13 to July 15.

Sixth Annual CHAI Workshop

08 Jun 2022

On the first weekend in June, CHAI held its 6th annual workshop in person at Asilomar Conference Grounds in Pacific Grove, CA.

CHAI Papers Published

04 Jun 2022

Here are some of the papers that have been published by CHAI students and affiliates recently: 

Evolving Curricula with Regret-Based Environment Design

12 May 2022

In this YouTube video, authors Jack Parker-Holder, Minqi Jiang, CHAI’s Michael Dennis, Mikayel Samvelyan, Jakob Foerster, Edward Grefenstette, and Tim Rocktäschel present ACCELL.

Blog Post: What’s Right and What’s Wrong with Optimizing for Engagement

02 May 2022

CHAI’s Jonathan Stray published a blog post with collaborators Pritanjana Bengani and Luke Thorburn.

Pieter Abbeel Receives 2021 ACM Prize in Computing

29 Apr 2022

On April 6, the Association for Computing Machinery recognized Pieter Abbeel as a pioneer of robot learning.

Facing Existential Risks through AI Proxies

10 Apr 2022

As a part of the Umea university #FrAIday seminar series about AI, Prof. Tom Lenaerts will present his work on understanding and modeling strategic decision-making in the face of existential risks that has been performed with students and international collaborators within his group in Brussels. The workhorse of this research is the collective risk dilemma (CRD), a formalization of the problem of climate negotiations, wherein the goal is to reach a common good, but selfish considerations may result in failure with potential cataclysmic outcomes. 

How to ensure safe development in a dynamic technology race? The difficulty of hitting the target.

31 Mar 2022

On March 29, the LAILEC Conference, co-located with the EUH4D Data Forum and the SAS Project Event, kicked off with parallel panel discussions.

What Does it Mean to Give Someone What They Want? The Nature of Preferences in Recommender Systems

12 Mar 2022

On March 11, Jonathan Stray, along with Luke Thorburn and Pri Bengani, published an article on Medium diving into the idea of “preferences”. The post explains that watching someone’s choices cannot tell us what they want, or what is best from them. CHAI’s Michael Dennis, Smitha Milli, Adam Gleave, and Micah Carroll also provided feedback during the development of the article.