#### Paper Group ANR 282

April 2, 2020

Few-Shot Learning with Geometric Constraints. Reinforcement Quantum Annealing: A Quantum-Assisted Learning Automata Approach. Being Bayesian, Even Just a Bit, Fixes Overconfidence in ReLU Networks. Crowdsourced Labeling for Worker-Task Specialization Block Model. Feature-Robustness, Flatness and Generalization Error for Deep Neural Networks. When R …