An Instance-Dependent Simulation Framework for Learning with Label Noise

1 code implementation • 23 Jul 2021 • Keren Gu, Xander Masotto, Vandana Bachani, Balaji Lakshminarayanan, Jack Nikodem, Dong Yin

We propose a simulation framework for generating instance-dependent noisy labels via a pseudo-labeling paradigm.

Learning with noisy labels

Using Videos to Evaluate Image Model Robustness

no code implementations • 22 Apr 2019 • Keren Gu, Brandon Yang, Jiquan Ngiam, Quoc Le, Jonathon Shlens

Compared to previous studies on adversarial examples and synthetic distortions, natural robustness captures a more diverse set of common image transformations that occur in the natural environment.

Efficient Model Learning for Human-Robot Collaborative Tasks

no code implementations • 24 May 2014 • Stefanos Nikolaidis, Keren Gu, Ramya Ramakrishnan, Julie Shah

We present a framework for learning human user models from joint-action demonstrations that enables the robot to compute a robust policy for a collaborative task with a human.

Clustering