scholarly journals Knowledge Distillation with Adversarial Samples Supporting Decision Boundary

2019 ◽  
Vol 33 ◽  
pp. 3771-3778 ◽  
Author(s):  
Byeongho Heo ◽  
Minsik Lee ◽  
Sangdoo Yun ◽  
Jin Young Choi
Keyword(s):  
Open Problem ◽  
Accurate Information ◽  
Decision Boundary ◽  
Good Decision ◽  
The Arts ◽  
Knowledge Distillation ◽  
Good Technique ◽  
Adversarial Attack ◽  
New Perspective ◽  
Trained Network

Many recent works on knowledge distillation have provided ways to transfer the knowledge of a trained network for improving the learning process of a new one, but finding a good technique for knowledge distillation is still an open problem. In this paper, we provide a new perspective based on a decision boundary, which is one of the most important component of a classifier. The generalization performance of a classifier is closely related to the adequacy of its decision boundary, so a good classifier bears a good decision boundary. Therefore, transferring information closely related to the decision boundary can be a good attempt for knowledge distillation. To realize this goal, we utilize an adversarial attack to discover samples supporting a decision boundary. Based on this idea, to transfer more accurate information about the decision boundary, the proposed algorithm trains a student classifier based on the adversarial samples supporting the decision boundary. Experiments show that the proposed method indeed improves knowledge distillation and achieves the state-of-the-arts performance.

scholarly journals Android ANDROID BASED AUGMENTED REALITY FOR DIGITAL VISUAL NAVIGATION

2019 ◽  
Vol 7 (2) ◽  
pp. 67
Author(s):  
Ronny Juwono
Keyword(s):  
Augmented Reality ◽  
Management System ◽  
Visual Navigation ◽  
Accurate Information ◽  
Android Application ◽  
Logistic Management ◽  
Online Business ◽  
New Perspective ◽  
Future Work ◽  
Map Service

Augmented Reality (AR) used in online business not merely to give new perspective of customer experience to product. AR is also used in operational such as in logistic to deliver product to the customer. The online map service such as Google Map, sometime fails to bring real visualization to user, especially when they are in unfamiliar places. AR may provide visualization of certain location’s coordinate. This digital visual navigation will help the user to choose which direction they should go, especially when Google Maps do not give accurate information related to the user’s surrounding. This papper provide solution to create Android application that can generate Augmented Reality navigation while user is using Google Map to travel. This papper is aimed to support the navigation for logistic management system especially for online business companies. In spite of the result of this reasearch, some findings and issues need to be discussed in the future. There was delay to generate the AR objects especially when user moves the device around fast. A simple 2D AR object is suggested for future work in order to decrease the delay.

scholarly journals Suspicion-Free Adversarial Attacks on Clustering Algorithms

2020 ◽  
Vol 34 (04) ◽  
pp. 3625-3632
Author(s):  
Anshuman Chhabra ◽  
Abhishek Roy ◽  
Prasant Mohapatra
Keyword(s):  
Machine Learning ◽  
Supervised Learning ◽  
Clustering Algorithms ◽  
Black Box ◽  
Single Sample ◽  
Decision Boundary ◽  
Empirical Results ◽  
Adversarial Attack ◽  
Modern Machine

Clustering algorithms are used in a large number of applications and play an important role in modern machine learning– yet, adversarial attacks on clustering algorithms seem to be broadly overlooked unlike supervised learning. In this paper, we seek to bridge this gap by proposing a black-box adversarial attack for clustering models for linearly separable clusters. Our attack works by perturbing a single sample close to the decision boundary, which leads to the misclustering of multiple unperturbed samples, named spill-over adversarial samples. We theoretically show the existence of such adversarial samples for the K-Means clustering. Our attack is especially strong as (1) we ensure the perturbed sample is not an outlier, hence not detectable, and (2) the exact metric used for clustering is not known to the attacker. We theoretically justify that the attack can indeed be successful without the knowledge of the true metric. We conclude by providing empirical results on a number of datasets, and clustering algorithms. To the best of our knowledge, this is the first work that generates spill-over adversarial samples without the knowledge of the true metric ensuring that the perturbed sample is not an outlier, and theoretically proves the above.

scholarly journals From Design Thinking to Art Thinking with an Open Innovation Perspective—A Case Study of How Art Thinking Rescued a Cultural Institution in Dublin

2018 ◽  
Vol 4 (4) ◽  
pp. 57 ◽  
Author(s):  
Peter Robbins
Keyword(s):  
Design Thinking ◽  
Single Case ◽  
Deep Understanding ◽  
Innovation Process ◽  
Optimal Position ◽  
Innovation Activities ◽  
The Arts ◽  
Current Scenario ◽  
New Perspective

This article uses a contemporary and revelatory case study to explore the relationship between three conversations in the innovation literature: Design Thinking, creativity in strategy, and the emerging area of Art Thinking. Businesses are increasingly operating in a VUCA environment where they need to design better experiences for their customers and better outcomes for their firm and the Arts are no exception. Innovation, or more correctly, growth through innovation, is a top priority for business and although there is no single, unifying blueprint for success at innovation, Design Thinking is the process that is receiving most attention and getting most traction. We review the literature on Design Thinking, showing how it teaches businesses to think with the creativity and intuition of a designer to show a deep understanding of, and have empathy with, the user. However, Design Thinking has limitations. By placing the consumer at the very heart of the innovation process, Design Thinking can often lead to more incremental, rather than radical, ideas. Now there is a new perspective emerging, Art Thinking, in which the objective is not to design a journey from the current scenario, A, to an improved position, A+. Art Thinking requires the creation of an optimal position B, and spends more time in the open-ended problem space, staking out possibilities and looking for uncontested space. This paper offers a single case study of a national arts organisation in Dublin facing an existential crisis, which used an Art Thinking approach successfully to give a much-needed shot in the arm to its commercial innovation activities.

scholarly journals Novel Model Based on Stacked Autoencoders with Sample-Wise Strategy for Fault Diagnosis

2019 ◽  
Vol 2019 ◽  
pp. 1-10
Author(s):  
Diehao Kong ◽  
Xuefeng Yan
Keyword(s):  
Fault Diagnosis ◽  
Chemical Engineering ◽  
Ground Truth ◽  
Student Model ◽  
Teacher Student ◽  
Stacked Autoencoders ◽  
Knowledge Distillation ◽  
New Perspective ◽  
Current Student ◽  
Teacher Model

Autoencoders are used for fault diagnosis in chemical engineering. To improve their performance, experts have paid close attention to regularized strategies and the creation of new and effective cost functions. However, existing methods are modified on the basis of only one model. This study provides a new perspective for strengthening the fault diagnosis model, which attempts to gain useful information from a model (teacher model) and applies it to a new model (student model). It pretrains the teacher model by fitting ground truth labels and then uses a sample-wise strategy to transfer knowledge from the teacher model. Finally, the knowledge and the ground truth labels are used to train the student model that is identical to the teacher model in terms of structure. The current student model is then used as the teacher of next student model. After step-by-step teacher-student reconfiguration and training, the optimal model is selected for fault diagnosis. Besides, knowledge distillation is applied in training procedures. The proposed method is applied to several benchmarked problems to prove its effectiveness.

Knowledge distillation based on decision boundary instances generated by DBI-GAN

2021 ◽  
Author(s):  
Ziqi Zhu ◽  
Xi Liu ◽  
Chunhua Deng ◽  
Jing Liu ◽  
Jixin Zou
Keyword(s):  
Decision Boundary ◽  
Knowledge Distillation

scholarly journals Improved Knowledge Distillation via Teacher Assistant

2020 ◽  
Vol 34 (04) ◽  
pp. 5191-5198 ◽  
Author(s):  
Seyed Iman Mirzadeh ◽  
Mehrdad Farajtabar ◽  
Ang Li ◽  
Nir Levine ◽  
Akihiro Matsukawa ◽  
...  
Keyword(s):  
Neural Networks ◽  
Theoretical Analysis ◽  
Network Performance ◽  
Deep Neural Networks ◽  
Sensor Nodes ◽  
Popular Method ◽  
Knowledge Distillation ◽  
Teacher Assistant ◽  
Trained Network

Despite the fact that deep neural networks are powerful models and achieve appealing results on many tasks, they are too large to be deployed on edge devices like smartphones or embedded sensor nodes. There have been efforts to compress these networks, and a popular method is knowledge distillation, where a large (teacher) pre-trained network is used to train a smaller (student) network. However, in this paper, we show that the student network performance degrades when the gap between student and teacher is large. Given a fixed student network, one cannot employ an arbitrarily large teacher, or in other words, a teacher can effectively transfer its knowledge to students up to a certain size, not smaller. To alleviate this shortcoming, we introduce multi-step knowledge distillation, which employs an intermediate-sized network (teacher assistant) to bridge the gap between the student and the teacher. Moreover, we study the effect of teacher assistant size and extend the framework to multi-step distillation. Theoretical analysis and extensive experiments on CIFAR-10,100 and ImageNet datasets and on CNN and ResNet architectures substantiate the effectiveness of our proposed approach.

scholarly journals Adversarial Attack for Uncertainty Estimation: Identifying Critical Regions in Neural Networks

2021 ◽  
Author(s):  
Ismail Alarab ◽  
Simant Prakoonwit
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Model Uncertainty ◽  
Uncertainty Estimation ◽  
Decision Boundary ◽  
Uncertainty Estimates ◽  
Data Points ◽  
Novel Method ◽  
Adversarial Attack ◽  
Critical Regions

AbstractWe propose a novel method to capture data points near decision boundary in neural network that are often referred to a specific type of uncertainty. In our approach, we sought to perform uncertainty estimation based on the idea of adversarial attack method. In this paper, uncertainty estimates are derived from the input perturbations, unlike previous studies that provide perturbations on the model's parameters as in Bayesian approach. We are able to produce uncertainty with couple of perturbations on the inputs. Interestingly, we apply the proposed method to datasets derived from blockchain. We compare the performance of model uncertainty with the most recent uncertainty methods. We show that the proposed method has revealed a significant outperformance over other methods and provided less risk to capture model uncertainty in machine learning.

scholarly journals Chinese and North American culture: A new perspective in linguistics studies

2019 ◽  
Vol 3 (1) ◽  
pp. 14-31
Author(s):  
Ziang Xiu Li ◽  
Cheng Yu Huan
Keyword(s):  
North America ◽  
North American ◽  
Chinese Culture ◽  
American Culture ◽  
The United States ◽  
The Arts ◽  
The People ◽  
The Two Cultures ◽  
American Way ◽  
New Perspective

We explored the two cultures in the two countries. There has been discussed on Chinese culture and North American culture. Chinese language, ceramics, architecture, music, dance, literature, martial arts, cuisine, visual arts, philosophy, business etiquette, religion, politics, and history have global influence, while its traditions and festivals are also celebrated, instilled, and practiced by people around the world. The culture of North America refers to the arts and other manifestations of human activities and achievements from the continent of North America. The American way of life or simply the American way is the unique lifestyle of the people of the United States of America. It refers to a nationalist ethos that adheres to the principle of life, liberty and the pursuit of happiness.

scholarly journals Analisa dan Perancangan Sistem Informasi Aktiva Tetap dalam Upaya Pengendalian aktiva Tetap yang Bersumber dari Pembelian dan Donasi

2011 ◽  
Vol 2 (2) ◽  
pp. 853
Author(s):  
Eka Novianti
Keyword(s):  
Decision Making ◽  
Information System ◽  
Accurate Information ◽  
User Requirements ◽  
Management Information ◽  
Important Work ◽  
Good Decision ◽  
Financial Position ◽  
Capital Assets ◽  
Fixed Assets

In an institution of capital assets is one of the very important work that needs to be managed well, both in terms of maximizing the value of benefits and terms of existence that can be used optimally. Likewise, management information about the assets that will be used for good decision-making. A system designed to optimally utilize the information. Fixed Assets Information System that meets the user requirements and works properly can provide precise and accurate information is required. Users can easily find out the value of the acquisition and the value of accumulated depreciation of assets which no longer belongs to the company. The values that have an impact on the Financial Position Report will generate a report more accurate and reliable. 

Sign in / Sign up

Export Citation Format

Share Document