- Dynamics in Truth Learning
- Project Year:
2024
- REU Student (s):
Julia Krizanova | Charles University (Prague, Czech Republic)
- Student 1 Institution:
Charles University (Prague, Czech Republic)
- Project Mentor:
Jie Gao
- Project Mentor Area:
Computer Science
- Project Abstract:
Consider a network N consisting of n agents, where all of them with ability to learn, want to correctly determine the value, i.e. state of the world they live in. Knowledge of each of the agents consists of his own private information and of actions that the other agents in the agent's neighborhood made before him. The actions of agents are being made in a sequential setting following an ordering σ. The goal is to achieve so called asymptotic truth learning on the whole network, meaning that almost all agents make a correct prediction of the current state of the world. To illustrate, when the number of agents n → ∞, then the probability of predicting the correct ground truth approaches one. In this paper we investigate and dive further into the potential relation of the truth learning and the field of statistical physics, and aim to perceive the problem from the dynamics point of view.
