Showing 1–4 of 4 results for author: Deri, J

The NVIDIA PilotNet Experiments

Authors: Mariusz Bojarski, Chenyi Chen, Joyjit Daw, Alperen Değirmenci, Joya Deri, Bernhard Firner, Beat Flepp, Sachin Gogri, Jesse Hong, Lawrence Jackel, Zhenhua Jia, BJ Lee, Bo Liu, Fei Liu, Urs Muller, Samuel Payne, Nischal Kota Nagendra Prasad, Artem Provodin, John Roach, Timur Rvachov, Neha Tadimeti, Jesper van Engelen, Haiguang Wen, Eric Yang, Zongyi Yang

Abstract: Four years ago, an experimental system known as PilotNet became the first NVIDIA system to steer an autonomous car along a roadway. This system represents a departure from the classical approach for self-driving in which the process is manually decomposed into a series of modules, each performing a different task. In PilotNet, on the other hand, a single deep neural network (DNN) takes pixels as i… ▽ More Four years ago, an experimental system known as PilotNet became the first NVIDIA system to steer an autonomous car along a roadway. This system represents a departure from the classical approach for self-driving in which the process is manually decomposed into a series of modules, each performing a different task. In PilotNet, on the other hand, a single deep neural network (DNN) takes pixels as input and produces a desired vehicle trajectory as output; there are no distinct internal modules connected by human-designed interfaces. We believe that handcrafted interfaces ultimately limit performance by restricting information flow through the system and that a learned approach, in combination with other artificial intelligence systems that add redundancy, will lead to better overall performing systems. We continue to conduct research toward that goal. This document describes the PilotNet lane-kee** effort, carried out over the past five years by our NVIDIA PilotNet group in Holmdel, New Jersey. Here we present a snapshot of system status in mid-2020 and highlight some of the work done by the PilotNet group. △ Less

Submitted 17 October, 2020; originally announced October 2020.

Graph Equivalence Classes for Spectral Projector-Based Graph Fourier Transforms

Authors: Joya A. Deri, José M. F. Moura

Abstract: We define and discuss the utility of two equivalence graph classes over which a spectral projector-based graph Fourier transform is equivalent: isomorphic equivalence classes and Jordan equivalence classes. Isomorphic equivalence classes show that the transform is equivalent up to a permutation on the node labels. Jordan equivalence classes permit identical transforms over graphs of nonidentical t… ▽ More We define and discuss the utility of two equivalence graph classes over which a spectral projector-based graph Fourier transform is equivalent: isomorphic equivalence classes and Jordan equivalence classes. Isomorphic equivalence classes show that the transform is equivalent up to a permutation on the node labels. Jordan equivalence classes permit identical transforms over graphs of nonidentical topologies and allow a basis-invariant characterization of total variation orderings of the spectral components. Methods to exploit these classes to reduce computation time of the transform as well as limitations are discussed. △ Less

Submitted 11 January, 2017; originally announced January 2017.

Agile Inexact Methods for Spectral Projector-Based Graph Fourier Transforms

Authors: Joya A. Deri, José M. F. Moura

Abstract: We propose an inexact method for the graph Fourier transform of a graph signal, as defined by the signal decomposition over the Jordan subspaces of the graph adjacency matrix. This method projects the signal over the generalized eigenspaces of the adjacency matrix, which accelerates the transform computation over large, sparse, and directed adjacency matrices. The trade-off between execution time… ▽ More We propose an inexact method for the graph Fourier transform of a graph signal, as defined by the signal decomposition over the Jordan subspaces of the graph adjacency matrix. This method projects the signal over the generalized eigenspaces of the adjacency matrix, which accelerates the transform computation over large, sparse, and directed adjacency matrices. The trade-off between execution time and fidelity to the original graph structure is discussed. In addition, properties such as a generalized Parseval's identity and total variation ordering of the generalized eigenspaces are discussed. The method is applied to 2010-2013 NYC taxi trip data to identify traffic hotspots on the Manhattan grid. Our results show that identical highly expressed geolocations can be identified with the inexact method and the method based on eigenvector projections, while reducing computation time by a factor of 26,000 and reducing energy dispersal among the spectral components corresponding to the multiple zero eigenvalue. △ Less

Submitted 11 January, 2017; originally announced January 2017.

Spectral Projector-Based Graph Fourier Transforms

Authors: Joya A. Deri, José M. F. Moura

Abstract: The paper presents the graph Fourier transform (GFT) of a signal in terms of its spectral decomposition over the Jordan subspaces of the graph adjacency matrix $A$. This representation is unique and coordinate free, and it leads to unambiguous definition of the spectral components ("harmonics") of a graph signal. This is particularly meaningful when $A$ has repeated eigenvalues, and it is very use… ▽ More The paper presents the graph Fourier transform (GFT) of a signal in terms of its spectral decomposition over the Jordan subspaces of the graph adjacency matrix $A$. This representation is unique and coordinate free, and it leads to unambiguous definition of the spectral components ("harmonics") of a graph signal. This is particularly meaningful when $A$ has repeated eigenvalues, and it is very useful when $A$ is defective or not diagonalizable (as it may be the case with directed graphs). Many real world large sparse graphs have defective adjacency matrices. We present properties of the GFT and show it to satisfy a generalized Parseval inequality and to admit a total variation ordering of the spectral components. We express the GFT in terms of spectral projectors and present an illustrative example for a real world large urban traffic dataset. △ Less

Submitted 10 January, 2017; originally announced January 2017.