Summary of Inverse Decision-making Using Neural Amortized Bayesian Actors, by Dominik Straub et al.
Inverse decision-making using neural amortized Bayesian actors
by Dominik Straub, Tobias F. Niehues, Jan Peters, Constantin A. Rothkopf
First submitted to arxiv on: 4 Sep 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Neurons and Cognition (q-bio.NC); Machine Learning (stat.ML)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary Bayesian observer and actor models have been widely used in cognitive science and neuroscience to provide normative explanations for various behavioral phenomena. These models attribute variability and biases to interpretable entities such as perceptual and motor uncertainty, prior beliefs, and behavioral costs. However, when extending these models to more naturalistic tasks with continuous actions, solving the Bayesian decision-making problem becomes analytically intractable. The inverse decision-making problem, i.e., performing inference over the parameters of such models given behavioral data, is even more computationally challenging. To overcome these limitations, the authors propose amortizing a Bayesian actor using a neural network trained on a wide range of parameter settings in an unsupervised fashion. This pre-trained neural network enables efficient gradient-based Bayesian inference of the Bayesian actor model’s parameters. The authors demonstrate their method’s effectiveness on synthetic data and empirical data from three sensorimotor tasks, showing that it can explain individuals’ behavioral patterns. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Bayesian observer and actor models are used to understand how our brains work when we make decisions or actions. These models try to figure out why we might make different choices or have different reactions. However, these models get really complicated when they’re applied to real-life situations where there are many possible actions. The researchers found a way to simplify this process by using a special kind of computer program called a neural network. This program can help us understand why people behave in certain ways and how we can explain those behaviors. |
Keywords
» Artificial intelligence » Bayesian inference » Inference » Neural network » Synthetic data » Unsupervised
