World Machine - Data & Report - Toy1D - Experiment 1 Configuration Test

Building on the previous experiment, Experiment 0 Protocol Test, which demonstrated the importance of the training protocol for the model's final performance, we investigate in this experiment the specific impact of each protocol step and several model architecture elements. 

World Machine is a research project that investigates the concept and creation of computational world models. These AI systems create internal representations to understand and make predictions about the external world. See the project page for more information. The project is part of the H.IAAC, the Hub for Artificial Intelligence and Cognitive Architecture, located at the Universidade Estadual de Campinas (UNICAMP), Brazil.

The files in this registry are organized by file extension. Each extension contains:

• json: metrics and logits metadata
• md: metrics
• memmap: logits data
• pt: trained models
• png: plots
• svg: figures with experiment pipelines
• final_results: specific files of the final results
• txt: verification files
• bin: experimental variation parameters