Schedule#
Week 1: Basics#
| Time | Mon, Sep 22 | Tue, Sep 23 | Wed, Sep 24 | Thu, Sep 25 | Fri, Sep 26 |
|---|---|---|---|---|---|
| 16:30 | To be announced (Nathaniel Daw) | Automated Experimental Design With SweetPea (Matthew Flatt) | Against theory-motivated experimentation in science (Marina Dubova) | ||
| 17:00 | Introduction to Automated Scientific Discovery (Sebastian Musslick) | Constructing and Fitting Computational Models of Mind and Brain With PsyNeuLink (Jonathan D. Cohen) | SweetPea (Younes Strittmatter) | Automated Experiment Sampling (Younes Strittmatter) | |
| 17:30 | Coffee | ||||
| 18:00 | PsyNeuLink (Younes Strittmatter) | SweetBean (Younes Strittmatter) | Integrative Experimental Design (Tom Griffiths) | ||
| 18:30 | |||||
| 19:00 | Dinner | ||||
| 19:30 | |||||
| 20:00 | Introduction to AutoRA (Younes Strittmatter) | Integrating PsyNeuLink with AutoRA (Younes Strittmatter) | Integrating Web-Based Experiments with AutoRA (Younes Strittmatter) | Closing the Loop: Integrating Samplers (Younes Strittmatter) | |
| 20:30 |
Week 2: Model Discovery and LLMs#
Sebastian Musslick : Introduction to Automated Scientific Discovery#
Readings:
{‘title’: ‘AutoRA Paper’, ‘url’: ‘https://psyarxiv.com/3y5bk/’}
Abstract: Short abstract here.
Nathaniel Daw : To be announced#
Abstract: Short abstract here.
Jonathan D. Cohen : Constructing and Fitting Computational Models of Mind and Brain With PsyNeuLink#
Abstract: Short abstract here.
Matthew Flatt : Automated Experimental Design With SweetPea#
Abstract: Short abstract here
Marina Dubova : Against theory-motivated experimentation in science#
Abstract: Short abstract here.
Younes Strittmatter : Automated Experiment Sampling#
Slides: PowerPoint
Abstract: Short abstract here.
Tom Griffiths : Integrative Experimental Design#
Abstract: Short abstract here.
Nathan Kutz : Sparse Identification of Non-Linear Dynamics#
Abstract: Sensing is a universal task in science and engineering. Downstream tasks from sensing include learning dynamical models, inferring full state estimates of a system (system identification), control decisions, and forecasting. These tasks are exceptionally challenging to achieve with limited sensors, noisy measurements, and corrupt or missing data. Existing techniques typically use current (static) sensor measurements to perform such tasks and require principled sensor placement or an abundance of randomly placed sensors. In contrast, we propose a SHallow REcurrent Decoder (SHRED) neural network structure which incorporates (i) a recurrent neural network (LSTM) to learn a latent representation of the temporal dynamics through sparse identification of nonlinear dynamics, and (ii) a shallow decoder that learns a mapping between this latent representation and the high-dimensional state space. By explicitly accounting for the time-history, or trajectory, of the sensor measurements, SHRED enables accurate reconstructions with far fewer sensors, learns parsimonious dynamical models, outperforms existing techniques when more measurements are available, and is agnostic towards sensor placement. In addition, a compressed representation of the high-dimensional state is directly obtained from sensor measurements, which provides an on-the-fly compression for modeling physical and engineering systems. Forecasting is also achieved from the sensor time-series data alone, producing an efficient paradigm for predicting temporal evolution with an exceptionally limited number of sensors. In the example cases explored, including turbulent flows, complex spatio-temporal dynamics can be characterized with exceedingly limited sensors that can be randomly placed with minimal loss of performance.
Marcello Mattar : Knowledge Distillation with Tiny-RNNs#
Abstract: Short abstract here.
Kevin Miller : Cognitive model discovery via disentangled RNNs#
Abstract: Short abstract here.
Kyle LaFollette : Equation Discovery in Reinforcement Learning#
Abstract: Short abstract here.
Daniel Weinhardt : Sparse Identification of Cognitive Equations#
Abstract: Short abstract here.
Kevin Miller : Discovering symbolic cognitive models from behavior#
Abstract: Short abstract here.
Milena Rmus, Akshay Kumar Jagadish · Milena Rmus, Akshay Kumar Jagadish : Generating computational cognitive models using LLMs#
Abstract: Short abstract here.
Victoria Bosch : CorText: From visually evoked brain responses to text captions#
Abstract: Short abstract here.
Marcel Binz : Centaur: A foundation Model of Cognition#
Abstract: Short abstract here.
Peter Clark : Multi-Agent LLMs for Scientific Discovery#
Abstract: Short abstract here.
Martyna Plomecka : LLMs for Automated Datascience#
Abstract: Short abstract here.