Erin Grant
University of Alberta
Overview
Analytical Connectionism is a 2-week summer course on analytical tools, including methods from statistical physics and probability theory, for probing neural networks and higher-level cognition. The course brings together neuroscience, psychology and machine-learning communities, and introduces attendees to analytical methods for neural-network analysis and connectionist theories of higher-level cognition and psychology.
Connectionism, a key theoretical approach in psychology, uses neural-network models to simulate a wide range of phenomena, including perception, memory, decision-making, language, and cognitive control. However, most connectionist models remain, to a certain extent, black boxes, and we lack a mathematical understanding of their behaviours. Recent progress in theoretical neuroscience and machine learning has provided novel analytical tools that have advanced our mathematical understanding of deep neural networks, and have the potential to help make these “black boxes” more transparent.
During the School, teams of students work closely to develop research projects on topics related to analytical connectionism, presenting initial proposals during week one and interim results at the School’s conclusion. Participant projects from prior Schools have led to publications at NeurIPS. Furthermore, students collaboratively produce lecture notes, which will then be collected in a peer-reviewed special issue to create a permanent resource for the community. Participant projects and notes from prior schools have led to publications at venues like NeurIPS and PMLR.
Additionally, students are grouped based on their expertise and preferences and assigned to take notes for a specific lecturer. These notes are peer-reviewed and published in a special journal issue. Currently, lecture notes from the 2023 and 2024 editions of the school are being collected into a publication in the Proceedings of Machine Learning Research (PMLR). This initiative aims to make the content accessible to future participants and those who were unable to attend, while providing note-takers with the opportunity to contribute to a formal publication.
📚 This year's School will have a topical focus on "Reinforcement Learning and Sequential Decision-Making."
This is the mathematical framework for how an agent learns to make a sequence of decisions in a dynamic environment, a process central to how humans and animals interact with the world. We will use powerful new analytical tools from fields like physics and math to build theories of sequential decision-making, with the ultimate goal of developing models that not only predict behavior but also provide profound new insights into the nature of intelligence.
This is the mathematical framework for how an agent learns to make a sequence of decisions in a dynamic environment, a process central to how humans and animals interact with the world. We will use powerful new analytical tools from fields like physics and math to build theories of sequential decision-making, with the ultimate goal of developing models that not only predict behavior but also provide profound new insights into the nature of intelligence.
This course will introduce:
- mathematical methods for neural-network analysis, providing a solid overview of the analytical tools available to understand neural-network models;
- key connectionist models and empirical findings from developmental psycholinguistics, which provide targets for analytical results.
During the course, you will:
- attend lectures given by leading researchers on theoretical methods and applications, key connectionist models, and experimental observations;
- participate in tutorials, poster sessions, and contributed talks.
- present to and engage with lecturers, organisers, and other participants during a poster session;
- collaborate in a group with other participants on a novel research project, mentored by the course organisers and lecturers.
- work together and co-author lecture notes for publication in a special issue.
Important dates
All dates are to be intended anywhere on earth time (AoE).
- Applications open:
- February 5, 2027
- Application deadline:
- April 16, 2027
- Outcome communicated:
- May 23, 2027
- Deadline to accept admission:
- June 13, 2027
Application details
Applications to participate in the 2027 School on Analytical Connectionism are now closed.
Target audience
This course is appropriate for graduate students, postdoctoral fellows, and early-career faculty in psychology, neuroscience, physics, computer science, and mathematics. The course is limited to 40 attendees to ensure a balance across fields. Attendees are expected to have a strong background in one of these disciplines and to have made some effort to introduce themselves to a complementary discipline.
We are committed to making this course open to all talented individuals. We strongly encourage students from underrepresented groups to apply and will offer bursaries to applicants who face financial barriers. In circumstances where all other things are equal, priority will be given to qualified applicants from underrepresented groups.
Course fees
A course fee of 200 CAD is required upon acceptance. This fee helps cover administrative and material costs for the two-week program.
What’s Provided:
- Week 1 (UAlberta): Lunch, coffee/tea breaks on course days, and one course dinner.
- Week 2 (BIRS): On-site accommodation and all meals are provided by BIRS.
Attendees are responsible for their own travel expenses, accommodation during the first week in Edmonton, and any other personal costs.
Cancellation policy:
- Within 7 days of registering/making the booking: 100% refund
- After 7 days of registering/making the booking and up to 15 days before the start of the school: 80% refund
- Less than 15 days before the start of the school: no refund
Financial assistance via travel grant may be available for successful applicants who find it difficult to take up a place for financial reasons. Applicants are asked to indicate in their application if they would like to be considered for financial aid. The amount of financial aid available will depend on the course funding from grants and sponsors.
Course Content
This year’s school will unite mathematicians, theoretical physicists, and statisticians with leading computational and experimental cognitive scientists to focus on reinforcement learning and sequential decision-making. The curriculum is designed to move beyond descriptive models of cognition toward mathematically-grounded, explanatory theories.
The workshop’s technical focus will be on the following analytical methods:
- Dynamical mean-field theory: A powerful method for reducing the study of high-dimensional stochastic differential equations, which describe learning, to a low-dimensional, self-consistent analysis.
- Optimal control theory in high dimensions: Exploring recent advances using statistical physics to make the principles of optimal control computationally tractable for systems with millions of parameters.
- Markov decision processes: The core mathematical framework for reinforcement learning, grounding problems of decision-making in the theories of stochastic processes and dynamic programming.
- High-dimensional probability and statistics: The study of concentration of measure and random matrix theory, which are foundational for understanding why large neural networks generalize.
Organizers
FAQ
- I need a visa to attend the programme. Will you provide support? If you require a visa to enter Canada, we will provide an official invitation letter to support your application after you have accepted your place. Please begin the visa application process as soon as possible, as processing times can be long.
- Where can I stay in Edmonton and Banff? Both locations offer a range of accommodation options. For the first week in Edmonton, university residences may be available. For the second week in Banff, BIRS will provide accommodation to all selected participants. More specific recommendations will be provided to accepted attendees.
- Will I receive individual feedback if my application is not successful? Unfortunately, due to the large number of applicants, we are unable to offer individual feedback on unsuccessful applications.
- What are the requirements for poster presentations? Due to space constraints, posters must be in A1 portrait format (594mm x 841mm).
- What are the requirements for spotlight presentations? Presentations will be short, likely a maximum of 5 minutes. We will email final guidelines to all presenters closer to the programme start date.