Home | Events

03

Jun

Teaser image to Explainable Methods for Reinforcement Learning

 

Explainable Methods for Reinforcement Learning

Jasmina Gajcin, Trinity College Dublin

   03.06.2024

   4:15 pm - 5:45 pm

   LMU Department of Statistics and via zoom

Deep reinforcement learning (DRL) algorithms have been successfully devel- oped for many high-risk real-life tasks in many fields such as autonomous driving, healthcare and finance. However, these algorithms rely on neural networks, making their decisions difficult to understand and interpret.

In this talk, I will cover some of the main challenges for developing explainable DRL methods, especially focusing on the difference between supervised and reinforcement learning from the perspective of explainability. Additionally, a part of this talk will be focused on counterfactual explanations in RL. Counterfactual explanations are a powerful explanation method and can explain outcomes by contrasting them with similar events which led to a different outcome. The talk will delve into how counterfactual explanations can be utilized in an RL setting.

Related

Link to Privacy, Data Privacy, and Differential Privacy

 

Colloquium  •  16.07.2024  •  LMU Department of Statistics and via zoom

Privacy, Data Privacy, and Differential Privacy

Colloquium at the LMU Department of Statistics with James Bailie from Harvard University.

Link to Variational Learning for Large Deep Networks

 

Colloquium  •  10.07.2024  •  LMU Department of Statistics and via zoom

Variational Learning for Large Deep Networks

Colloquium at the LMU Department of Statistics with Thomas Möllenhoff from RIKEN, Tokyo.

Link to Can today’s intention to treat have a causal effect on tomorrow’s hazard function?

 

Colloquium  •  03.07.2024  •  LMU Department of Statistics and via zoom

Can today’s intention to treat have a causal effect on tomorrow’s hazard function?

Colloquium at the LMU Department of Statistics with Jan Beyersmann, University of Ulm.

Link to The Complexities of Differential Privacy for Survey Data

 

Colloquium  •  26.06.2024  •  LMU Department of Statistics and via zoom

The Complexities of Differential Privacy for Survey Data

Link to Resampling-based inference for the average treatment effect in observational studies with competing risks

 

Colloquium  •  19.06.2024  •  LMU Department of Statistics and via zoom

Resampling-based inference for the average treatment effect in observational studies with competing risks

This talk explores three resampling methods to construct valid confidence intervals and bands for treatment effect estimation in competing risks studies.