Cooperative AI Events

Cooperative AI Summer School

Aims and Focus
The Cooperative AI Summer School is designed to provide students and early-career professionals in AI, computer science, and related disciplines with a firm grounding in the emerging field of cooperative AI.

The curriculum for the summer school will range from the "foundations" to the "frontiers" of the field, with lectures delivered by those at the forefront of cooperative AI research. The foundational aspects of the program will explore the objectives and key concepts of cooperative AI, delving into the theoretical underpinnings of the main challenges and opportunities associated with improving the cooperative intelligence of advanced AI systems. Participants will also be exposed to cutting-edge developments on the frontiers of cooperative AI research, which build upon recent progress in areas such as language modelling and multi-agent reinforcement learning.

In addition, the summer school will provide the opportunity for participants to share their own work, and to build connections with others in, or entering, the field. The Cooperative AI Foundation is committed to the growth of a diverse and inclusive research community, and we especially welcome applications from underrepresented backgrounds. Further information about financial assistance for attendees can be found below.

Recordings of the lectures from the summer school can be found on the Cooperative AI Foundation's YouTube channel.

Date
16–19 July 2023
Location
London, UK
Cost
Free / £249 / £499
Deadline
7 May 2023
Notification
15 May 2023
Application

Speakers

Noam Brown
Researcher, OpenAI

Noam Brown is a Researcher at OpenAI. Before that, he worked on multi-agent artificial intelligence at Facebook AI Research, with a particular focus on imperfect-information games. He co-created Libratus and Pluribus, the first AIs to defeat top humans in two-player no-limit poker and multiplayer no-limit poker, respectively. He has received the Marvin Minsky Medal for Outstanding Achievements in AI, was named one of MIT Tech Review's 35 Innovators Under 35, and his work on Pluribus was named by Science Magazine to be one of the top 10 scientific breakthroughs of the year. Noam received his PhD from Carnegie Mellon University, where he received the School of Computer Science Distinguished Dissertation Award.

Vincent Conitzer
Professor, Carnegie Mellon University
Professor, University of Oxford

Vincent Conitzer is Professor of Computer Science (with affiliate/courtesy appointments in Machine Learning, Philosophy, and the Tepper School of Business) at Carnegie Mellon University, where he directs the Foundations of Cooperative AI Lab (FOCAL). He is also Head of Technical AI Engagement at the Institute for Ethics in AI, and Professor of Computer Science and Philosophy, at the University of Oxford. Previous to joining CMU, Conitzer was the Kimberly J. Jenkins Distinguished University Professor of New Technologies and Professor of Computer Science, Professor of Economics, and Professor of Philosophy at Duke University. He received Ph.D. (2006) and M.S. (2003) degrees in Computer Science from Carnegie Mellon University, and an A.B. (2001) degree in Applied Mathematics from Harvard University.

Yali Du
Assistant Professor, King's College London

Yali Du is a Lecturer (Assistant Professor) at King’s College London. Her research aim is to enable machines to exhibit cooperative and trusted behaviour in intelligent decision making tasks.  Her research interest lies in reinforcement learning and evaluation, multi-agent cooperation, and social agency (e.g.  human-involved learning, safety and ethics). She was chosen for AAAI New Faculty Highlights program (2023). Prior to her current position, Yali received her PhD from the University of Technology Sydney and worked as a postdoc at University College London.

Jakob Foerster
Associate Professor, University of Oxford

Jakob Foerster is an Associate Professor at the department of engineering science at the University of Oxford. During his PhD at Oxford he helped bring deep multi-agent reinforcement learning to the forefront of AI research and interned at Google Brain, OpenAI, and DeepMind. After his PhD he worked as a research scientist at Facebook AI Research in California, where he continued doing foundational work. He was the lead organizer of the first Emergent Communication workshop at NeurIPS in 2017, which he has helped organize ever since and was awarded a prestigious CIFAR AI chair in 2019.

Edward Hughes
Staff Research Engineer, DeepMind

Edward Hughes is a Staff Research Engineer at DeepMind. His research pioneers the field of Cooperative AI, algorithms that work in partnership with each other and humans. His work lies at a crossroads linking multi-agent reinforcement learning with sociology and economics.

Edward received his PhD in Theoretical Physics from Queen Mary University of London on applications of string theory to particle scattering. He read Mathematics at Cambridge University, where he graduated with Distinction.

Edward’s publications are available on Google Scholar.

Natasha Jaques
Assistant Professor, University of Washington
Senior Research Scientist, Google Brain

Natasha Jaques is an Assistant Professor at the University of Washington and a Senior Research Scientist at Google Brain. Her research focuses on Social Reinforcement Learning in multi-agent and human-AI interactions. Natasha completed her PhD at the MIT Media Lab, where her thesis received the Outstanding PhD Dissertation Award from the Association for the Advancement of Affective Computing, and completed a postdoc at UC Berkeley. Her work has received Best Demo at NeurIPS, an honourable mention for Best Paper at ICML, Best of Collection in the IEEE Transactions on Affective Computing, and received several best paper awards at NeurIPS and AAAI workshops. She has interned at DeepMind, Google Brain, and was an OpenAI Scholars mentor. Her work has been featured in Science Magazine, MIT Technology Review, Quartz, IEEE Spectrum, Boston Magazine, and on CBC radio. Natasha earned her Masters degree from the University of British Columbia, and undergraduate degrees in Computer Science and Psychology from the University of Regina.

Joel Leibo
Senior Staff Research Scientist, DeepMind

Joel Z. Leibo is a research Scientist at DeepMind. He obtained his PhD in 2013 from MIT where he worked on the computational neuroscience of face recognition with Tomaso Poggio. Nowadays, Joel's research is aimed at the following questions: How can we get deep reinforcement learning agents to perform complex cognitive behaviors like cooperating with one another in groups? How should we evaluate the performance of deep reinforcement learning agents? How can we model processes like cumulative culture that gave rise to unique aspects of human intelligence?

Nisarg Shah
Assistant Professor, University of Toronto

Nisarg Shah is an Associate Professor of computer science at the University of Toronto. He has been recognized as part of "Innovators Under 35" by MIT Technology Review Asia Pacific in 2022 and "AI's 10 to Watch" by IEEE Intelligent Systems in 2020. He is also the winner of the 2016 IFAAMAS Victor Lesser Distinguished Dissertation Award and the 2014-2015 Facebook PhD Fellowship. Shah conducts research at the intersection of computer science and economics, addressing issues of fairness, efficiency, elicitation, and incentives that arise when humans are affected by algorithmic decision-making. His recent work develops theoretical foundations for fairness in fields such as voting, resource allocation, and machine learning. He has co-developed two not-for-profit websites, Spliddit.org and RoboVote.org, which have helped more than 200,000 users make provably fair and optimal decisions in their everyday lives. He earned his PhD in computer science at Carnegie Mellon University and was a postdoctoral fellow at Harvard University.

Schedule

Programme Details

The majority of the summer school will take place between Monday 17 July and Wednesday 19 July, beginning with a welcome reception on the evening of Sunday 16 July. Participants are welcome to arrive from 5pm onwards on Sunday; the event will end on Wednesday afternoon.

Schedule

Monday 17 July 2023

Time

Event

Speakers

09:30am
Introduction to Cooperative AI

Lewis Hammond (University of Oxford / Cooperative AI Foundation)

10:00am
Fostering Cooperation via Fairness in AI Systems

Nisarg Shah (University of Toronto)

11:30am
A Foundation Model for Cooperative AI

Edward Hughes (Google DeepMind)

02:00pm
Foundations of Cooperative AI

Vincent Conitzer (Carnegie Mellon University, University of Oxford)

03:30pm
Aligning AI to Everyone via Reinforcement Learning

Yali Du (King's College London)

04:30pm
RLHF: How to Learn from Human Feedback with Reinforcement Learning (with Applications to Cooperative AI)

Natasha Jaques (University of Washington, Google DeepMind)

05:30pm
Poster Session 1

Tuesday 18 July 2023

Time

Event

Speakers

09:30am
Risks and Threat Models

Akbir Khan (University College London / Cooperative AI Foundation)

10:00am
Learning to Compete and Cooperate via Self-Play

Noam Brown (Open AI)

11:30am
A Theory of Appropriateness with Applications to Generative AI

Joel Leibo (Google DeepMind)

2:00pm
Opponent Shaping and Interference in General-Sum Games

Jakob Foerster (University of Oxford)

03:30pm
Practical 1: Interactions Between Language Models in ChatArena

Zhengyao Jiang (University College London)

03:30pm
Practical 2: Open-Source Game Theory

Casper Oesterheld (Carnegie Mellon University)

05:30pm
Poster Session 2

Wednesday 19 July 2023

Time

Event

Speakers

10:30am

Jesse Clifton (Center on Long Term Risk / Cooperative AI Foundation)

A Cooperative AI Workflow
11.00am

Michael Dennis (Google DeepMInd)

Unsupervised Environment Design in Multi-Agent Reinforcement Learning and its Implications for Cooperative AI
12:00pm

Lewis Hammond (University of Oxford / Cooperative AI Foundation)

Closing Remarks

Application

Please fill out this form to apply for your place at the Cooperative AI Summer School by 23:59 AoE (UTC-12) on 7 May 2023. Applications from underrepresented backgrounds are especially welcome. Notification of acceptance will be provided by 15 May 2023 at the latest. If accepted, you will be invited to confirm your registration by responding with further details and making any required payments.

The cost of the summer school is £249 for students and independent researchers, and £499 for faculty and other working professionals. This includes a hotel room for three nights (Sunday, Monday, Tuesday), dinner on the first night, and all breakfasts and lunches. There is no additional tax.

CAIF is committed to ensuring that no-one who wishes to attend the summer school is prevented from doing so due to a lack of funding. If you do not have an institution or employer that is able to cover the registration fee, you will have the opportunity to request your fee to be waived. You will be required to upload a letter of support from your academic supervisor or equivalent. If you would not be able to attend without additional financial support (to cover your flights, for example), please contact us, and we will also be happy to discuss this with you.

Sponsors

Cooperative AI Foundation

The mission of the Cooperative AI Foundation is to support research that will improve the cooperative intelligence of advanced AI for the benefit of all.

Visit sponsor website

FAQs

Will the summer school be in-person, virtual, or hybrid? Will there be recordings?

The summer school is in-person only, although recordings will be available on the Cooperative AI Foundation's YouTube channel.

Who is eligible for this summer school?

Anyone (over the age of 18) can apply to the summer school, and we especially encourage applications from under-represented backgrounds. We expect all applicants to have at least a basic grounding in AI and the relevant mathematical background to contemporary AI research.

How will you decide whether or not I am accepted?

If the summer school is oversubscribed, we will prioritise those who: provide the strongest evidence of their interest in cooperative AI; are already pursuing graduate studies or working in a relevant field; come from under-represented backgrounds.

Will I receive any ECTS points or certification for participating?

The Cooperative AI Foundation is unable to provide any kind of formal academic credit for participating in the summer school. We are, however, able to provide a digital certificate of participation for all attendees, which your institution may be willing to accept as a form of academic credit.

What if I need a travel visa?

Depending on your nationality, you might need a visa to travel to the UK. This would be a short-stay, single-, or multiple-entry visa. Please check visit www.gov.uk/check-uk-visa to see if you need a UK visa. If you are accepted to the summer school and you need a visa, please contact us and we will send you an invitation letter.

What payment methods do you accept?

If you are accepted, you will be sent a unique link to pay online via Wise (formerly TransferWise).

What is your cancellation policy?

Cancellations at least 30 days before the start of the summer school will be refunded 50% of the registration fee, but cancellations less than 30 days before the start of the summer school will not be entitled to receive any refund. If it is possible to fill your place with another applicant, we may, at our discretion, fully reimburse you.

Can I apply to have my fees waived, or for any other financial support?

CAIF is committed to ensuring that no-one who wants to attend is prevented from doing so due to a lack of funding. Further details of the support we offer can be found above. Please contact us if you have any questions regarding this.

What kinds of costs will I need to pay myself or get reimbursed separately?

Travel to and from the summer school; dinners on the second and third nights (Monday and Tuesday); visa application costs; and other travel expenses. Please also see the answer to the previous question.

I have a question that isn’t answered here. How do I contact you?

Please use our contact form.