New Approach for Advanced Cooperative Learning Algorithms using RL Methods (ACLA)

Abstract

We explore a new approach for dynamic products availability in a three retailer shops in the market. Retailers can cooperate with each other and can get benefit from cooperative information by their own policies that accurately represent their goals and interests. The retailers are the learning agents in the system and use RL to learn cooperatively from the environment. The system becomes Markov decision process model on the basis of logical theory on the seller's inventory policy, the arrival process of the customers and refill times. Cooperation in learning (CL) can be understood in a multiagent system. The agents are capable of learning from both their own trials and other agents' knowledge. In this paper, we proposed a new approach for Advanced Cooperative Learning Algorithms using RL methods (ACLA). We have shown the performance comparison between cooperative learning algorithms and advanced cooperative learning algorithms using RL method with expertness measure. Expertness measuring criteria which were used in earlier work is further enhanced & improved in proposed method. Four methods for measuring the agents' expertness are used i.e. Normal (Nrm), Absolute (Abs), Positive (P), Negative (N). The novelty of this approach lies in the implementation of the RL algorithms with expertness measuring criteria by means of Sarsa learning and Sarsa(λ) learning algorithms. The paper shows implementation results and performance comparison of all these algorithms.