deep neural networks

Model Predictive Control and Imitation Learning Algorithms for Robot Motion Planning in Physical Human-Robot Interaction

-

B9 R2325

This seminar presents a framework for safe and efficient human-robot workspace sharing by using Deep Neural Networks (DNN) and safety filters to rapidly imitate computationally heavy Nonlinear Model Predictive Control method (NMPC), with successful experimental validation on a UR5 manipulator.

Approximation and Optimization for Neural Networks

-
In this talk, we consider new connections between the approximation and optimization of neural networks. Instead of relying on excessive over-parametrization to achieve zero training loss, we identify good minima by comparison with established approximation bounds.

On training algorithms for neural networks

-

B9 L2 R2322 H1

In this talk, I will first give a convergence analysis of gradient descent (GD) method for training neural networks by relating them with finite element method. I will then present some acceleration techniques for GD method and also give some alternative training algorithms

Towards Designing Robust Deep Learning Models for 3D Understanding

-

B3 L5 R5220

Deep Neural Networks (DNNs) have shown huge success over the years to solve many 2D computer vision tasks driven by massive labeled 2D datasets and advancements in 2D vision models, but less success is witnessed on 3D vision tasks. This dissertation proposes innovative approaches to enhance the robustness of DNNs for 3D understanding and in 3D settings. The research focuses on two main areas: adversarial robustness on 3D data and setups, and the robustness of DNNs to realistic 3D scenarios. Two paradigms for 3D understanding are discussed: representing 3D as a set of 3D points and performing 2D processing of multiple images of the 3D data.