Summary of Pedestrian Crossing Decisions Can Be Explained by Bounded Optimal Decision-making Under Noisy Visual Perception, By Yueyang Wang et al.
Pedestrian crossing decisions can be explained by bounded optimal decision-making under noisy visual perception
by Yueyang Wang, Aravinda Ramakrishnan Srinivasan, Jussi P.P. Jokinen, Antti Oulasvirta, Gustav Markkula
First submitted to arxiv on: 6 Feb 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 presents a machine learning model that simulates pedestrian crossing decisions based on the theory of computational rationality. The model assumes that pedestrians make boundedly optimal choices due to cognitive limitations, combining both mechanistic and machine learning approaches. Specifically, it uses reinforcement learning to learn a bounded optimal behavior policy while modeling noisy human visual perception and assumed rewards in crossing. The results reproduce known empirical phenomena, including the effects of vehicle speed on gap acceptance and pedestrian timing, and suggest that behaviors previously framed as biases might be rational adaptations to constraints of visual perception. By leveraging both reinforcement learning and mechanistic modeling, this model offers novel insights into pedestrian behavior and may provide a foundation for more accurate and scalable models. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper is about a computer program that helps us understand how people decide whether to cross the street or not. The program uses two different ways of thinking: one that tries to figure out what people are thinking, and another that learns from trial and error. This combination of approaches helps the program make better predictions about when people will cross the street. It also shows that things we thought were strange behaviors might actually be the result of people making smart choices based on their own limitations. Overall, this research helps us understand how people make decisions, which can help us build safer and more efficient streets. |
Keywords
» Artificial intelligence » Machine learning » Reinforcement learning
