Summary of Deep Reinforcement Multi-agent Learning Framework For Information Gathering with Local Gaussian Processes For Water Monitoring, by Samuel Yanes Luis et al.
Deep Reinforcement Multi-agent Learning framework for Information Gathering with Local Gaussian Processes for Water Monitoring
by Samuel Yanes Luis, Dmitriy Shutin, Juan Marchal Gómez, Daniel Gutiérrez Reina, Sergio Toral Marín
First submitted to arxiv on: 9 Jan 2024
Categories
- Main: Artificial Intelligence (cs.AI)
- Secondary: Machine Learning (cs.LG)
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 authors propose a multi-agent system of autonomous surface vehicles to efficiently monitor water quality. They use Local Gaussian Processes and Deep Reinforcement Learning to jointly obtain effective monitoring policies. The proposed approach utilizes an information gain reward-based Deep Convolutional Policy, which makes decisions based on the mean and variance of the model. Training is done using Double Deep Q-Learning with a Consensus-based heuristic to minimize estimation error in a safe manner. Simulation results show an improvement of up to 24% in terms of mean absolute error, as well as smaller average estimation errors for monitoring water quality variables and algae blooms compared to state-of-the-art approaches. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Autonomous surface vehicles can help monitor water quality by collecting data on contamination levels. The paper proposes using a multi-agent system with these vehicles to efficiently monitor the water. They use special models, called Local Gaussian Processes, which are better at capturing information about different parts of the water. This helps the vehicles make good decisions based on what they see. Training the agents is done in a way that keeps them safe and accurate. The results show that this approach can be up to 24% more accurate than other methods. |
Keywords
* Artificial intelligence * Reinforcement learning
