Summary of Infinite Ends From Finite Samples: Open-ended Goal Inference As Top-down Bayesian Filtering Of Bottom-up Proposals, by Tan Zhi-xuan et al.
Infinite Ends from Finite Samples: Open-Ended Goal Inference as Top-Down Bayesian Filtering of Bottom-Up Proposals
by Tan Zhi-Xuan, Gloria Kang, Vikash Mansinghka, Joshua B. Tenenbaum
First submitted to arxiv on: 23 Jul 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 The paper introduces a sequential Monte Carlo model for open-ended goal inference, combining Bayesian inverse planning with sampling based on co-occurring subgoals. The model proposes goal hypotheses related to achieved subgoals without exhaustive search, then filters out irrational goals. Validation is done in the Block Words task, comparing against heuristic guessing and exact Bayesian inference over hundreds of goals. Results show that the model better predicts human goal inferences in terms of mean, variance, efficiency, and resource rationality, achieving similar accuracy at a fraction of cognitive cost. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper tries to understand how we can quickly figure out what other people are trying to achieve, even when there are many possibilities. The researchers created a model that combines two approaches: one that looks at the big picture and another that focuses on small details. They tested this model by asking people to guess the word someone is building with lettered blocks. The results show that their model does a better job of predicting what people will think than other methods, while also explaining some strange patterns in human behavior. |
Keywords
» Artificial intelligence » Bayesian inference » Inference
