Article Source

• Improving equality and productivity with AI-driven tax policies

Improving equality and productivity with AI-driven tax policies

ABSTRACT

Tackling real-world socio-economic challenges requires designing and testing economic policies, such as income tax policy. However, this is hard in practice, due to a lack of appropriate (micro-level) economic data and limited opportunity to experiment.

Our speaker, Stephan Zheng and his team at Salesforce Research, propose a two-level deep reinforcement learning approach to learn dynamic tax policies, based on principled economic simulations in which both agents and a social planner (government) learn and adapt.

This talk covers three main topics:

1. Economic policy design in the context of multi-agent RL.
2. Salesforce Research’s two-level RL approach to economic policy design.
3. Open research problems towards an AI Economist for the real world. These include key methodological challenges in two-level RL and data-driven economic modeling, multi-agent RL, mechanism design, robustness, explainability, and others.

Our speaker, Stephan Zheng, leads the AI Economist team at Salesforce Research. He works on using deep reinforcement learning and economic simulations to design economic policy. This research has been covered by the Financial Times, Forbes, MIT Tech Review, and others.