Adversariality in Machine Learning Systems: On Neural Networks and the Limits of Knowledge

2020 ◽  
pp. 197-225
Author(s):  
Théo Lepage-Richer
Keyword(s):  
Machine Learning ◽  
Neural Networks ◽  
Learning Systems ◽  
Limits Of Knowledge

INCREMENTAL LEARNING IN BIOLOGICAL AND MACHINE LEARNING SYSTEMS

2002 ◽  
Vol 12 (06) ◽  
pp. 447-465 ◽  
Author(s):  
STEPHAN K. CHALUP
Keyword(s):  
Machine Learning ◽  
Neural Networks ◽  
Artificial Neural Networks ◽  
Side Effects ◽  
Incremental Learning ◽  
Learning Systems ◽  
Experimental Results ◽  
Recurrent Artificial Neural Networks ◽  
Standard Learning

Incremental learning concepts are reviewed in machine learning and neurobiology. They are identified in evolution, neurodevelopment and learning. A timeline of qualitative axon, neuron and synapse development summarizes the review on neurodevelopment. A discussion of experimental results on data incremental learning with recurrent artificial neural networks reveals that incremental learning often seems to be more efficient or powerful than standard learning but can produce unexpected side effects. A characterization of incremental learning is proposed which takes the elaborated biological and machine learning concepts into account.

scholarly journals Adversarial Perturbation Defense on Deep Neural Networks

2022 ◽  
Vol 54 (8) ◽  
pp. 1-36
Author(s):  
Xingwei Zhang ◽  
Xiaolong Zheng ◽  
Wenji Mao
Keyword(s):  
Machine Learning ◽  
Neural Networks ◽  
High Probability ◽  
Deep Neural Networks ◽  
State Of The Art ◽  
Learning Systems ◽  
Small Perturbations ◽  
Image Objects ◽  
Comprehensive Survey ◽  
Human Eyes

Deep neural networks (DNNs) have been verified to be easily attacked by well-designed adversarial perturbations. Image objects with small perturbations that are imperceptible to human eyes can induce DNN-based image class classifiers towards making erroneous predictions with high probability. Adversarial perturbations can also fool real-world machine learning systems and transfer between different architectures and datasets. Recently, defense methods against adversarial perturbations have become a hot topic and attracted much attention. A large number of works have been put forward to defend against adversarial perturbations, enhancing DNN robustness against potential attacks, or interpreting the origin of adversarial perturbations. In this article, we provide a comprehensive survey on classical and state-of-the-art defense methods by illuminating their main concepts, in-depth algorithms, and fundamental hypotheses regarding the origin of adversarial perturbations. In addition, we further discuss potential directions of this domain for future researchers.

scholarly journals MARGIN: Uncovering Deep Neural Networks Using Graph Signal Analysis

2021 ◽  
Vol 4 ◽  
Author(s):  
Rushil Anirudh ◽  
Jayaraman J. Thiagarajan ◽  
Rahul Sridhar ◽  
Peer-Timo Bremer
Keyword(s):  
Machine Learning ◽  
Neural Networks ◽  
Signal Analysis ◽  
Deep Neural Networks ◽  
Large Family ◽  
Learning Systems ◽  
Large Set ◽  
Crucial Aspect ◽  
Building Trust ◽  
Influential Nodes

Interpretability has emerged as a crucial aspect of building trust in machine learning systems, aimed at providing insights into the working of complex neural networks that are otherwise opaque to a user. There are a plethora of existing solutions addressing various aspects of interpretability ranging from identifying prototypical samples in a dataset to explaining image predictions or explaining mis-classifications. While all of these diverse techniques address seemingly different aspects of interpretability, we hypothesize that a large family of interepretability tasks are variants of the same central problem which is identifying relative change in a model’s prediction. This paper introduces MARGIN, a simple yet general approach to address a large set of interpretability tasks MARGIN exploits ideas rooted in graph signal analysis to determine influential nodes in a graph, which are defined as those nodes that maximally describe a function defined on the graph. By carefully defining task-specific graphs and functions, we demonstrate that MARGIN outperforms existing approaches in a number of disparate interpretability challenges.

Robust Machine Learning Systems: Reliability and Security for Deep Neural Networks

2018 ◽  
Author(s):  
Muhammad Abdullah Hanif ◽  
Faiq Khalid ◽  
Rachmad Vidya Wicaksana Putra ◽  
Semeen Rehman ◽  
Muhammad Shafique
Keyword(s):  
Machine Learning ◽  
Neural Networks ◽  
Deep Neural Networks ◽  
Learning Systems

Ethical Rationality in AI

2021 ◽  
pp. 69-84
Author(s):  
Jonas Holst
Keyword(s):  
Machine Learning ◽  
Neural Networks ◽  
Artificial Neural Networks ◽  
Virtue Ethics ◽  
Black Box ◽  
Learning Systems ◽  
Ethical Problems ◽  
Starting Point ◽  
Full Ethical Agents ◽  
Philosophical Understanding

Taking its starting point in a discussion of the concept of intelligence, the chapter develops a philosophical understanding of ethical rationality and discusses its role and implications for two ethical problems within AI: Firstly, the so-called “black box problem,” which is widely discussed in the AI community, and secondly, another more complex one which will be addressed as the “Tin Man problem.” The first problem has to do with opacity, bias, and explainability in the design and development of advanced machine learning systems, such as artificial neural networks, whereas the second problem is more directly associated with the prospect for humans and AI of becoming full ethical agents. Based on Aristotelian virtue ethics, it will be argued that intelligence in human and artificial forms should approximate ethical rationality, which entails a well-balanced synthesis of reason and emotion.

scholarly journals An Empirical Model For Validity And Verification Of Ai Behavior: Overcoming Ai Hazards In Neural Networks

2021 ◽  
Vol 21 ◽  
pp. 44-52
Author(s):  
Ayse K Arslan
Keyword(s):  
Artificial Intelligence ◽  
Machine Learning ◽  
Neural Networks ◽  
Empirical Model ◽  
Learning Systems ◽  
Rapid Progress ◽  
Research Directions ◽  
Harmful Behavior ◽  
High Level ◽  
Applications Of Ai

Rapid progress in machine learning and artificial intelligence (AI) has brought increasing attention to the potential impacts of AI technologies on society. This paper discusses hazards in machine learning systems, defined as unintended and harmful behavior that may emerge from poor design of real-world AI systems with a particular focus on ANN. The paper provides a review of previous work in these areas as well as suggesting research directions with a focus on relevance to cutting-edge AI systems with a focus on neural networks. Finally, the paper considers the high-level question of how to think most productively about the safety of forward-looking applications of AI.

scholarly journals Towards Intelligent E-Learning Systems: A Hybrid Model for Predicatingthe Learning Continuity in Iraqi Higher Education

Webology ◽  
2020 ◽  
Vol 17 (2) ◽  
pp. 172-188
Author(s):  
Mohammed K. Kadhim ◽  
Alia K. Hassan
Keyword(s):  
Higher Education ◽  
Machine Learning ◽  
Neural Networks ◽  
Hybrid Model ◽  
Educational Process ◽  
Learning Systems ◽  
Learning System ◽  
The Internet ◽  
Learning Methods ◽  
E Learning

E-Learning system gains a great attention in the past years; with advance of the internet and the information exchange techniques the importance to merge the traditional learning means with the internet-based learning methods became a must especially in Iraq, the Iraqi higher education is now coping with the new information and communication technologies and adopting a modern methods for upgrading their education and learning ways. There are great efforts to blend E-Learning systems with the educational process, in order to fulfill this purposes the proposed research is advancing E-Learning systems by suggesting a hybrid method that combines two Artificial Intelligence Techniques (AI) inside the design and the development of an intelligent E-Learning system for computer science department at university of technology. The utilization of Artificial Neural Networks algorithm (ANNs) especially Recurrent Neural Networks (RNN) is a way of implementing deep learning technique to predict the students' final out comes in virtual class room based on their grades and their learning behaviors. RNN is optimized by utilizing ADAM optimizer to lift the accuracy of the proposed algorithm, the dataset are gathered and processed to suite the education purposes and was divided into80% for training the model and 20% for testing the model, the results of the hybrid model are compared with other machine learning methods like Multi-Layer Perceptron (MLP), decision tree, naïve Bayesian, and random forest using WEKA environment, the results of the proposed model showed a promising accuracy when compared with the mentioned machine learning algorithms.

scholarly journals The Algorithms of Mindfulness

2021 ◽  
pp. 016224392110256
Author(s):  
Johannes Bruder
Keyword(s):  
Machine Learning ◽  
Neural Networks ◽  
Artificial Neural Networks ◽  
Information Overload ◽  
Learning Systems ◽  
Emotional Resilience ◽  
Creative Cognition ◽  
North Americans ◽  
Artificial Neural ◽  
Algorithmic Techniques

This paper analyzes notions and models of optimized cognition emerging at the intersections of psychology, neuroscience, and computing. What I somewhat polemically call the algorithms of mindfulness describes an ideal that determines algorithmic techniques of the self, geared at emotional resilience and creative cognition. A reframing of rest, exemplified in corporate mindfulness programs and the design of experimental artificial neural networks sits at the heart of this process. Mindfulness trainings provide cues as to this reframing, for they detail each in their own way how intermittent periods of rest are to be recruited to augment our cognitive capacities and combat the effects of stress and information overload. They typically rely on and co-opt neuroscience knowledge about what the brains of North Americans and Europeans do when we rest. Current designs for artificial neural networks draw on the same neuroscience research and incorporate coarse principles of cognition in brains to make machine learning systems more resilient and creative. These algorithmic techniques are primarily conceived to prevent psychopathologies where stress is considered the driving force of success. Against this backdrop, I ask how machine learning systems could be employed to unsettle the concept of pathological cognition itself.

Platform for Analysing and Encouraging Student Activity on Contest and E-learning Systems

2018 ◽  
Vol 12 ◽  
pp. 85-98
Author(s):  
Bojan Kostadinov ◽  
Mile Jovanov ◽  
Emil STANKOV
Keyword(s):  
Machine Learning ◽  
Data Collection ◽  
Educational Policy ◽  
Learning Systems ◽  
Data Sources ◽  
Or Education ◽  
Student Activity ◽  
The World ◽  
E Learning ◽  
Analyse Data

Data collection and machine learning are changing the world. Whether it is medicine, sports or education, companies and institutions are investing a lot of time and money in systems that gather, process and analyse data. Likewise, to improve competitiveness, a lot of countries are making changes to their educational policy by supporting STEM disciplines. Therefore, it’s important to put effort into using various data sources to help students succeed in STEM. In this paper, we present a platform that can analyse student’s activity on various contest and e-learning systems, combine and process the data, and then present it in various ways that are easy to understand. This in turn enables teachers and organizers to recognize talented and hardworking students, identify issues, and/or motivate students to practice and work on areas where they’re weaker.

MACHINE LEARNING FOR SOFTWARE

2019 ◽  
pp. 251-255
Author(s):  
Shafagat Mahmudova
Keyword(s):  
Machine Learning ◽  
Neural Networks ◽  
Soft Computing ◽  
Computing Technology

The study machine learning for software based on Soft Computing technology. It analyzes Soft Computing components. Their use in software, their advantages and challenges are studied. Machine learning and its features are highlighted. The functions and features of neural networks are clarified, and recommendations were given.

Sign in / Sign up

Export Citation Format

Share Document