2024 School on Analytical Connectionism

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 behaviors. 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.

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 with links to experimental observations, 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, Q&A sessions, and panel discussions;
• take part in networking activities such as poster sessions;
• work in groups on a novel research project, mentored by the course organizers and lecturers.

Applications to participate in the 2024 School on Analytical Connectionism are now closed.

Target audience

This course is appropriate for graduate students, postdoctoral fellows and early-career faculty in a number of fields, including psychology, neuroscience, physics, computer science, and mathematics. 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.

The course is limited to just under 40 attendees, who will be chosen to balance the representation of different fields. In circumstances where all other things are equal, priority will be given to applicants from populations underrepresented in the scientific workforce as defined by NIH, including but not limited to racially underrepresented individuals, women, individuals with disabilities, and individuals from disadvantaged backgrounds.

Course fees

There are no course fees, but attendees are expected to cover their own travel, visa expenses, and any meals not offered by the summer school. (Morning and afternoon coffee breaks and lunch will be provided Monday to Friday.) Accommodation in NYC for students not living in NYC and the surrounding areas will be provided by the school.

Travel grants inclusive of the above named personal expenses will be offered to individuals whose participation furthers the goal to promote diversity in systems and computational neuroscience, in particular among populations underrepresented in the scientific workforce as defined by NIH.

Core lecturers

Topic lecturers

Teaching assistants

Project mentors

The 2024 Edition of the Analytical Connectionism Summery School will focus on using analytical models to study connectionism and its application to cognition. Topics will include developmental psychology, particularly how cognitive functions evolve, and memory, both from neurobiological and cognitive neuroscience perspectives. The course will also explore large language models (LLMs) and their relation to language processing, alongside discussions in computational neuroscience on sensory processing and decision-making. Together, these areas will provide a thorough understanding of cognition through analytical and computational approaches.

Schedule

Contributed posters

1. Adam Manoogian “Contextual Inference Underlies Decision Making in Schizophrenia: An Active Inference Model”
2. Akif Erdem Sagtekin “Emergent excitatory/inhibitory balance in neural networks during task training”
3. Alessandro Favero “Hierarchies and Compositionality in Diffusion Models”
4. Ali Karami “Investigation of Numerosity Representation in Convolution Neural Networks”
5. Alisa Leshchenko “Specialization in a minimal task-trained network”
6. Anushri Arora “A New Look at Low Rank Recurrent Neural Networks”
7. Asit Pal “Multistage Recurrent Circuit Model IMplementing Normalization”
8. Bin Wang “Desegregation of Neuronal Predictive Processing”
9. Catherine Chen “Representations of Semantic Relations in the Human Brain”
10. Claudia Merger “Learning Interacting Theories from Data”
11. Dongyu Gong “Fundamental Limits in the Working Memory Capacity of Large Language Models”
12. Dota Dong “Multimodal Video Transformers Partially Align with Multimodal Grounding and Compositionality in the Brain”
13. Ganesh Kumar “Place Field Reirganization as State Representation Learning to Improve Policy Convergence”
14. Ji-An Li “Deep Learning without Weight Symmetry”
15. Jing Li “Dynamic self-efficacy as a computational mechanism of mania emergence”
16. Lindsay Smith “Learning Continous Chaotic Attractors with a reservoir Computer”
17. Mia Whitefield “The Generalisability and Flexibility of Representations Across Learning in Humans and Neural Networks”
18. Mildred Salgado-Menez “Characterization of neural correlates of Macaca mulatta hippocampus in a visual metronome task”
19. Nathan Cloos “Differentiable Optimization of Similarity Scores Between Models and Brains”
20. Po-Chen Kuo “Uncovering the Computation of Dynamic Foraging with Actor-Critic Recurrent Neural Networks”
21. Shawn Rhoads “Distinct effects of depression and social anxiety on social craving computations”
22. Shujun Xiong “Counting Intersections on Smooth Manifolds Bounds External Memory in Deterministic Network Activity”
23. Tobias Thomas “Modeling Dataset bias in machine-learned theories of economic decision-making”
24. Valentin Schmutz “High-dimensional neuronal activity from low-dimensional population dynamics:an exactly solvable model”
25. Veronica Chelu “Dual receptor model of serotonergic psychedelics”
26. Zachary Friedenberger “Dendritic excitability controls overdispersion”
27. Zihan Zhang Axonal “Dendritic Overlap Recurrent Neural Network”

Participant list

1. Abdel Mfougouon Njupoun
2. Adam Manoogian
3. Akif Erdem Sagtekin
4. Alessandro Favero
5. Ali Karami
6. Alisa Leshchenko
7. Anushri Arora
8. Asit Pal
9. Bin Wang
10. Catherine Chen
11. Claudia Merger
12. Conor McGrory
13. Dongyu Gong
14. Dota Dong
15. Ganesh Kumar
16. Huadong Xiong
17. Ilia Sucholutsky
18. Imran Thobani
19. Ji-An Li
20. Jing Li
21. Katya Ivshina
22. Lindsay Smith
23. Mia Whitefield
24. Mildred Salgado-Menez
25. Nathan Cloos
26. Po-Chen Kuo
27. Shawn Rhoads
28. Shujun Xiong
29. Tobias Thomas
30. Valentin Schmutz
31. Veronica Chelu
32. Zachary Friedenberger
33. Zihan Zhang

Course organizers