Honda motor co., ltd. (20240330651). DISCOVERING INTERPRETABLE DYNAMICALLY EVOLVING RELATIONS (DIDER) simplified abstract
Contents
DISCOVERING INTERPRETABLE DYNAMICALLY EVOLVING RELATIONS (DIDER)
Organization Name
Inventor(s)
Enna Sachdeva of San Jose CA (US)
Chiho Choi of San Jose CA (US)
DISCOVERING INTERPRETABLE DYNAMICALLY EVOLVING RELATIONS (DIDER) - A simplified explanation of the abstract
This abstract first appeared for US patent application 20240330651 titled 'DISCOVERING INTERPRETABLE DYNAMICALLY EVOLVING RELATIONS (DIDER)
The abstract describes a method for discovering interpretable dynamically evolving relations (DIDER) using a DIDER model for multi-agent interactions represented by edge embeddings.
- Training the DIDER model involves feeding edge embeddings to a Long Short-Term Memory (LSTM) network to generate forward and reverse outputs.
- The LSTM forward output is then fed to a duration encoder to generate an edge duration output.
- The DIDER model is trained based on a probability distribution for different edge types obtained from the LSTM outputs fed to an edge prior and an edge encoder.
Potential Applications: - Understanding complex interactions between multiple agents in various scenarios. - Predicting future trajectories and behaviors of agents in dynamic environments.
Problems Solved: - Providing a method to interpret and analyze evolving relations between agents. - Enhancing the understanding of multi-agent interactions in dynamic systems.
Benefits: - Improved insights into complex interactions between agents. - Better prediction of future behaviors and trajectories in dynamic environments.
Commercial Applications: - This technology could be applied in fields such as autonomous vehicles, robotics, and social network analysis to improve decision-making processes and optimize interactions between agents.
Questions about the technology: 1. How does the DIDER model improve upon existing methods for analyzing multi-agent interactions? 2. What are the potential limitations of using edge embeddings in training the DIDER model?
Frequently Updated Research: - Stay updated on advancements in LSTM networks and edge embedding techniques for analyzing multi-agent interactions.
Original Abstract Submitted
according to one aspect, discovering interpretable dynamically evolving relations (dider) may including using a dider model for multi-agent interactions represented by an execution set of edge embeddings indicative of trajectory interactions between two or more agents for one or more time steps. the dider model may be trained by feeding a training set of edge embeddings to a long short-term memory network (lstm) forward to generate an lstm forward output, feeding the training set of edge embeddings to a lstm reverse to generate an lstm reverse output, feeding the lstm forward output to a duration encoder to generate an edge duration output, and training the dider model based on a probability distribution for one or more different edge types obtained by feeding the lstm forward output or the lstm reverse output to an edge prior and an edge encoder.
