Summary of Goals As Reward-producing Programs, by Guy Davidson et al.
Goals as Reward-Producing Programs
by Guy Davidson, Graham Todd, Julian Togelius, Todd M. Gureckis, Brenden M. Lake
First submitted to arxiv on: 21 May 2024
Categories
- Main: Artificial Intelligence (cs.AI)
- Secondary: None
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 This paper bridges the gap between current goal-oriented models and the richness of everyday human goals by collecting a dataset of playful goals, modeling them as reward-producing programs, and generating novel human-like goals through program synthesis. The authors collect a dataset of scorable, single-player games and model them as symbolic operations that capture temporal constraints and allow for program execution on behavioral traces to evaluate progress. They learn a fitness function over the infinite set of possible goal programs and sample novel goals with a quality-diversity algorithm. Human evaluators found that model-generated goals were indistinguishable from human-created games, and the internal fitness scores predict games that are evaluated as more fun to play and more human-like. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This research helps computers better understand what people want to achieve in their daily lives. The team collected a big dataset of simple games that people like to play for fun. They then used this data to create new games that are similar to the ones humans like. The results showed that these new games are just as enjoyable and realistic as the ones created by humans themselves. This discovery can help us build more advanced computers that can understand what we want to achieve in our daily lives. |
