International Business Machines Corporation

Mastercard International Incorporated

Himanshi

Albee Jhoney

Akash V Gunjal

Neeraj Kumar Kashyap

Shiv Markam

Pushpendra Singh Dahiya

Akanksha Singh

Mridul Sayana

Embodiments provide methods and systems for reducing bias in an artificial intelligence model. A method includes computing, by a processor, a reward value based at least in part on a similarity between model predictions from a pre-trained model and agent predictions from a Reinforcement Learning (RL) agent. The method includes performing each step of one or more steps of a rule of a plurality of rules. The rule is assigned a weight and the rule includes a protected attribute, a cumulative statistic value type, and a comparison threshold. The method includes sending a cumulative reward value generated using the reward value and each weighted punishment value computed based at least in part on applying each rule of the plurality of rules to the RL agent. The RL agent learns to biases from the agent predictions while maintaining similarity with model predictions by maximizing the cumulative reward value.

Methods and systems for reducing bias in an artificial intelligence model

An approach is provided in which an information handling system captures a set of performance metrics corresponding to a traffic flow through a set of microservices included in the service mesh. Next, the information handling system identifies one or more modifications to the traffic flow based on analyzing the set of performance metrics against a set of application requirements corresponding to the set of microservices. In turn, the information handling system adjusts a traffic routing policy of the service mesh based on the identified one or more modifications.

Manage a network of microservices

Growing community of inventors

New Delhi, India

Himanshi