LCAD - Contribuições do(a) usuário(a) [pt-br]

Simulating Robotic Cars Using Time-Delay Neural Networks

2018-05-05T16:53:18Z

Jacson Silva: Criou página com 'category:Publicações A. F. De Souza and J. R. C. da Silva and F. Mutz and C. Badue and T. Oliveira-Santos Authors: [http://www.lcad.inf.ufes.br/team/index.php/Prof._Dr._...'

[[category:Publicações]]
A. F. De Souza and J. R. C. da Silva and F. Mutz and C. Badue and T. Oliveira-Santos
Authors: [http://www.lcad.inf.ufes.br/team/index.php/Prof._Dr._Alberto_Ferreira_De_Souza Alberto F. de Souza], [http://jeiks.net/ Jacson Rodrigues Correia-Silva], [http://fmtz.com.br/ Filipe Mutz], [https://www.inf.ufes.br/~claudine/ Claudine Badue], [http://www.inf.ufes.br/~todsantos/home Thiago Oliveira-Santos]

'''International Joint Conference on Neural Networks''': [https://doi.org/10.1109/IJCNN.2016.7727342]

==Abstract==

[[Arquivo:IARA.jpg]]

In this paper, we propose a simulator for robotic cars based on two time-delay neural networks. These networks are intended to simulate the mechanisms that govern how a set of effort commands changes the car's velocity and the direction it is moving. The first neural network receives as input a temporal sequence of current and previous throttle and brake efforts, along with a temporal sequence of the previous car's velocities (estimated by the network), and outputs the velocity that the real car would reach in the next time interval given these inputs. The second neural network estimates the arctangent of curvature (a variable related to the steering wheel angle) that a real car would reach in the next time interval given a temporal sequence of current and previous steering efforts and previous arctangents of curvatures of the car estimated by the network. We evaluated the performance of our simulator using real-world datasets acquired using an autonomous robotic car. Experimental results showed that our simulator was able to simulate in real time how a set of efforts influences the car's velocity and arctangent of curvature. While navigating in a map of a real-world environment, our car simulator was able to emulate the velocity and arctangent of curvature of the real car with mean squared error of 2.2×10-3 (m/s)2 and 4.0×10-5 rad2, respectively.

== Source-Code ==

Available [https://github.com/LCAD-UFES/dgfann_lcad here]

== BibTeX ==

@INPROCEEDINGS{souzaijcnn2016,
author = {A. F. De Souza and J. R. C. da Silva and F. Mutz and C. Badue and T. Oliveira-Santos},
booktitle = {2016 International Joint Conference on Neural Networks (IJCNN)},
title = {Simulating robotic cars using time-delay neural networks},
year = {2016},
pages = {1261-1268},
doi = {10.1109/IJCNN.2016.7727342},
month = {July}
}

Publicações do LCAD

2018-05-05T16:45:33Z

Jacson Silva:

[[category:Publicações]]
== [http://www.lcad.inf.ufes.br/~alberto/pesquisadores-lcad-producao Publicações do LCAD desde 1996] ==

== Papers Summaries ==

[[A Particle Filter-based Lane Marker Tracking Approach using a Cubic Spline Model]], 2015, SIBGRAPI

[[A Facial Expression Recognition System Using Convolutional Networks]], 2015, SIBGRAPI

[[Facial Expression Recognition with Convolutional Neural Networks: Coping with Few Data and the Training Sample Order]], 2016, Pattern Recognition International Journal

[[Simulating Robotic Cars Using Time-Delay Neural Networks]], 2016, International Joint Conference on Neural Networks

[[Automatic Large-Scale Data Acquisition via Crowdsourcing for Crosswalk Classification: A Deep Learning Approach]], 2017, Computers & Graphics

[[Deep Learning Based Large-Scale Automatic Satellite Crosswalk Classification]], 2017, IEEE Geoscience and Remote Sensing Letters

[[Ego-Lane Analysis System|Ego-Lane Analysis System (ELAS): Dataset and Algorithms]], 2017, Image and Vision Computing

[[Copycat CNN: Stealing Knowledge by Persuading Confession with Random Non-Labeled Data]], 2018, International Joint Conference on Neural Networks

Publicações do LCAD

2018-05-05T16:43:28Z

Jacson Silva:

[[category:Publicações]]
== [http://www.lcad.inf.ufes.br/~alberto/pesquisadores-lcad-producao Publicações do LCAD desde 1996] ==

== Papers Summaries ==

[[A Particle Filter-based Lane Marker Tracking Approach using a Cubic Spline Model]], 2015, SIBGRAPI

[[A Facial Expression Recognition System Using Convolutional Networks]], 2015, SIBGRAPI

[[Facial Expression Recognition with Convolutional Neural Networks: Coping with Few Data and the Training Sample Order]], 2016, Pattern Recognition International Journal

[[Automatic Large-Scale Data Acquisition via Crowdsourcing for Crosswalk Classification: A Deep Learning Approach]], 2017, Computers & Graphics

[[Deep Learning Based Large-Scale Automatic Satellite Crosswalk Classification]], 2017, IEEE Geoscience and Remote Sensing Letters

[[Ego-Lane Analysis System|Ego-Lane Analysis System (ELAS): Dataset and Algorithms]], 2017, Image and Vision Computing

[[Copycat CNN: Stealing Knowledge by Persuading Confession with Random Non-Labeled Data]], 2018, International Joint Conference on Neural Networks

Copycat CNN: Stealing Knowledge by Persuading Confession with Random Non-Labeled Data

2018-05-05T16:39:51Z

Jacson Silva: Criou página com 'category:Publicações Authors: [http://jeiks.net/ Jacson Rodrigues Correia-Silva], [http://rodrigoberriel.com/ Rodrigo Berriel], [https://www.inf.ufes.br/~claudine/ Claud...'

[[category:Publicações]]
Authors: [http://jeiks.net/ Jacson Rodrigues Correia-Silva], [http://rodrigoberriel.com/ Rodrigo Berriel], [https://www.inf.ufes.br/~claudine/ Claudine Badue], [http://www.lcad.inf.ufes.br/team/index.php/Prof._Dr._Alberto_Ferreira_De_Souza Alberto F. de Souza], [http://www.inf.ufes.br/~todsantos/home Thiago Oliveira-Santos]

'''International Joint Conference on Neural Networks'''

==Abstract==

[[Arquivo:copycat-fig-01.png]]

In the past few years, Convolutional Neural Networks (CNNs) have been achieving state-of-the-art performance on a variety of problems. Many companies employ resources and money to generate these models and provide them as an API, therefore it is in their best interest to protect them, i.e., to avoid that someone else copy them. Recent studies revealed that state-of-the-art CNNs are vulnerable to adversarial examples attacks, and this weakness indicates that CNNs do not need to operate in the problem domain (PD). Therefore, we hypothesize that they also do not need to be trained with examples of the PD in order to operate in it. Given these facts, in this paper, we investigate if a target black-box CNN can be copied by persuading it to confess its knowledge through random non-labeled data. The copy is twofold: i) the target network is queried with random data and its predictions are used to create a fake dataset with the knowledge of the network; and ii) a copycat network is trained with the fake dataset and should be able to achieve similar performance as the target network. This hypothesis was evaluated locally in three problems (facial expression, object, and crosswalk classification) and against a cloud-based API. In the copy attacks, images from both non-problem domain and PD were used. All copycat networks achieved at least 93.7% of the performance of the original models with non-problem domain data, and at least 98.6% using additional data from the PD. Additionally, the copycat CNN successfully copied at least 97.3% of the performance of the Microsoft Azure Emotion API. Our results show that it is possible to create a copycat CNN by simply querying a target network as black-box with random non-labeled data.

== Source-Code ==

Available [https://github.com/jeiks/Stealing_DL_Models here]

== BibTeX ==

@inproceedings{jacsonijcnn2018,
Author = {Jacson Rodrigues Correia-Silva, Rodrigo F. Berriel and Claudine Badue and Alberto F. de Souza and Thiago Oliveira-Santos},
Title = {Copycat CNN: Stealing Knowledge by Persuading Confession with Random Non-Labeled Data},
Journal = {International Joint Conference on Neural Networks},
booktitle = {2018 International Joint Conference on Neural Networks (IJCNN)},
year = {2018},
}