Marwen Belkaid

The paper introduces DeepLabSAM for accurate pose estimation in interacting mice and constructs the MTMB dataset, showing improved performance over existing methods.

This study compares two spiking neural models of dopamine-driven reinforcement learning, finding that the disinhibition model outperforms the direct control model in learning tasks.

This study shows that a humanoid robot's gaze affects human decision-making speed, decision thresholds, neural activity, and strategy choice in social interactions.

This study shows mice generate choice variability in a deterministic task through memory-free, pseudo-random selection rather than learning complex sequences.

This paper models uncertainty-seeking behavior mediated by cholinergic influence on dopamine neurons and shows it reproduces experimental data and predicts specific outcomes.