4 papers
On Transportability for Structural Causal Bandits
Min Woo Park, Sanghack Lee
Intelligent agents equipped with causal knowledge can optimize their action spaces to avoid unnecessary exploration. The structural causal bandit framework provides a graphical cha…
Mitigating Length Bias in RLHF through a Causal Lens
Hyeonji Kim, Sujeong Oh, Sanghack Lee
Reinforcement learning from human feedback (RLHF) is widely used to align large language models (LLMs) with human preferences. However, RLHF-trained reward models often exhibit len…
On Predicting Post-Click Conversion Rate via Counterfactual Inference
Junhyung Ahn, Sanghack Lee
Accurately predicting conversion rate (CVR) is essential in various recommendation domains such as online advertising systems and e-commerce. These systems utilize user interaction…
Toward a Complete Criterion for Value of Information in Insoluble Decision Problems
Ryan Carey, Sanghack Lee, Robin J. Evans
In a decision problem, observations are said to be material if they must be taken into account to perform optimally. Decision problems have an underlying (graphical) causal structu…