6 papers
Performative Prediction with Selective Labels
Giovani Valdrighi, Isabel Valera, Marcos Medeiros Raimundo
Many social applications of machine learning exhibit performative effects: population behavior changes in response to deployed models. Performative prediction studies this interact…
Learning Pareto Stationary Fronts via Single-Pass Backpropagation
Elina Rojin Celik, Marcos Medeiros Raimundo, Isabel Valera
We propose MOSEL (Multi-Objective Stackelberg Efficient Learning), a framework for a posteriori multi-objective optimization (MOO) in deep neural networks that recovers a full fron…
Beyond Routing Weights: Faithful Response-Level Interpretation of Mixture-of-Experts Reward Models via Contribution Contrast
Yifan Wang, Jinyi Mu, Mayank Jobanputra +4
Reward models are central to learning from human preferences, yet identifying what drives their predictions remains challenging. Recent sparse Mixture-of-Experts (MoE) reward model…
Sparse Mixture-of-Experts Reward Models Learn Interpretable and Specialized Experts for Personalized Preference Modeling
Yifan Wang, Jinyi Mu, Mayank Jobanputra +5
Preference modeling plays a central role in reinforcement learning from human feedback (RLHF), enabling large language models (LLMs) to align with human values. However, most exist…
Long-term Fairness with Selective Labels
Giovani Valdrighi, Isabel Valera, Marcos Medeiros Raimundo
Long-term fairness algorithms aim to satisfy fairness beyond static and short-term notions by accounting for the dynamics between decision-making policies and population behavior.…
Designing Long-term Group Fair Policies in Dynamical Systems
Miriam Rateike, Isabel Valera, Patrick Forré
Neglecting the effect that decisions have on individuals (and thus, on the underlying data distribution) when designing algorithmic decision-making policies may increase inequaliti…