scholarly journals The fractal geometry of fitness landscapes at the local optima level

2020 ◽  
Author(s):  
Sarah L. Thomson ◽  
Gabriela Ochoa ◽  
Sébastien Verel
Keyword(s):  
Fractal Dimension ◽  
Visual Analysis ◽  
Fitness Landscape ◽  
Fractal Dimensions ◽  
Machine Learning Techniques ◽  
Search Performance ◽  
Algorithm Performance ◽  
Local Optima ◽  
Learning Techniques ◽  
Insight Into

AbstractA local optima network (LON) encodes local optima connectivity in the fitness landscape of a combinatorial optimisation problem. Recently, LONs have been studied for their fractal dimension. Fractal dimension is a complexity index where a non-integer dimension can be assigned to a pattern. This paper investigates the fractal nature of LONs and how that nature relates to metaheuristic performance on the underlying problem. We use visual analysis, correlation analysis, and machine learning techniques to demonstrate that relationships exist and that fractal features of LONs can contribute to explaining and predicting algorithm performance. The results show that the extent of multifractality and high fractal dimensions in the LON can contribute in this way when placed in regression models with other predictors. Features are also individually correlated with search performance, and visual analysis of LONs shows insight into this relationship.

scholarly journals Machine Learning and Sensor Fusion for Estimating Continuous Energy Expenditure

AI Magazine ◽  
2012 ◽  
Vol 33 (2) ◽  
pp. 55 ◽  
Author(s):  
Nisarg Vyas ◽  
Jonathan Farringdon ◽  
David Andre ◽  
John Ivo Stivoric
Keyword(s):  
Machine Learning ◽  
Energy Expenditure ◽  
Sensor Data ◽  
Machine Learning Techniques ◽  
Sensor Technology ◽  
Health Goals ◽  
System A ◽  
Multi Sensor Data Fusion ◽  
Learning Techniques ◽  
Insight Into

In this article we provide insight into the BodyMedia FIT armband system — a wearable multi-sensor technology that continuously monitors physiological events related to energy expenditure for weight management using machine learning and data modeling methods. Since becoming commercially available in 2001, more than half a million users have used the system to track their physiological parameters and to achieve their individual health goals including weight-loss. We describe several challenges that arise in applying machine learning techniques to the health care domain and present various solutions utilized in the armband system. We demonstrate how machine learning and multi-sensor data fusion techniques are critical to the system’s success.

scholarly journals Research on threat detection in cyber security based on machine learning

2021 ◽  
Vol 2113 (1) ◽  
pp. 012074
Author(s):  
Qiwei Ke
Keyword(s):  
Machine Learning ◽  
Cyber Security ◽  
Information Privacy ◽  
Machine Learning Techniques ◽  
Threat Detection ◽  
Vulnerability Detection ◽  
New Information ◽  
Learning Techniques ◽  
Potential Problems ◽  
Insight Into

Abstract The volume of the data has been rocketed since the new information era arrives. How to protect information privacy and detect the threat whenever the intrusion happens has become a hot topic. In this essay, we are going to look into the latest machine learning techniques (including deep learning) which are applicable in intrusion detection, malware detection, and vulnerability detection. And the comparison between the traditional methods and novel methods will be demonstrated in detail. Specially, we would examine the whole experiment process of representative examples from recent research projects to give a better insight into how the models function and cooperate. In addition, some potential problems and improvements would be illustrated at the end of each section.

scholarly journals Inferring Future Landscapes: Sampling the Local Optima Level

2020 ◽  
Vol 28 (4) ◽  
pp. 621-641 ◽  
Author(s):  
Sarah L. Thomson ◽  
Gabriela Ochoa ◽  
Sébastien Verel ◽  
Nadarajen Veerapen
Keyword(s):  
Fitness Landscape ◽  
Quadratic Assignment Problem ◽  
Real Life ◽  
Quadratic Assignment ◽  
Search Performance ◽  
Local Optima ◽  
Sampling Algorithms ◽  
Local Optima Networks ◽  
Almost All ◽  
Construction Algorithms

Connection patterns among Local Optima Networks (LONs) can inform heuristic design for optimisation. LON research has predominantly required complete enumeration of a fitness landscape, thereby restricting analysis to problems diminutive in size compared to real-life situations. LON sampling algorithms are therefore important. In this article, we study LON construction algorithms for the Quadratic Assignment Problem (QAP). Using machine learning, we use estimated LON features to predict search performance for competitive heuristics used in the QAP domain. The results show that by using random forest regression, LON construction algorithms produce fitness landscape features which can explain almost all search variance. We find that LON samples better relate to search than enumerated LONs do. The importance of fitness levels of sampled LONs in search predictions is crystallised. Features from LONs produced by different algorithms are combined in predictions for the first time, with promising results for this “super-sampling”: a model to predict tabu search success explained 99% of variance. Arguments are made for the use-case of each LON algorithm and for combining the exploitative process of one with the exploratory optimisation of the other.

scholarly journals Machine Learning Techniques for Detecting Identifying Linguistic Patterns in News Media

2019 ◽  
Author(s):  
A Samuel Pottinger
Keyword(s):  
Machine Learning ◽  
News Media ◽  
Machine Learning Techniques ◽  
Linguistic Markers ◽  
Learning Techniques ◽  
Probabilistic Classifier ◽  
Linguistic Patterns ◽  
News Agencies ◽  
Filter Bubbles ◽  
Insight Into

An article's tone and framing not only influence an audience's perception of a story but may also reveal attributes of author identity and bias. Building upon prior media, psychological, and machine learning research, this neural network-based system detects those writing characteristics in ten news agencies' reporting, discovering patterns that, intentional or not, may reveal an agency's topical perspectives or common contextualization patterns. Specifically, learning linguistic markers of different organizations through a newly released open database, this probabilistic classifier predicts an article's publishing agency with 74% hidden test set accuracy given only a short snippet of text. The resulting model demonstrates how unintentional 'filter bubbles' can emerge in machine learning systems and, by comparing agencies' patterns and highlighting outlets' prototypical articles through an open source exemplar search engine, this paper offers new insight into news media bias.

scholarly journals Machine-learning a virus assembly fitness landscape

2019 ◽  
Author(s):  
Pierre-Philippe Dechant ◽  
Yang-Hui He
Keyword(s):  
Machine Learning ◽  
Fitness Landscape ◽  
Virus Assembly ◽  
Machine Learning Techniques ◽  
Whole Genome ◽  
Learning Techniques ◽  
Phenotype Space ◽  
Computationally Expensive ◽  
Stochastic Assembly ◽  
Evolutionary Fitness

AbstractRealistic evolutionary fitness landscapes are notoriously difficult to construct. A recent cutting-edge model of virus assembly consists of a dodecahedral capsid with 12 corresponding packaging signals in three affinity bands. This whole genome/phenotype space consisting of 312genomes has been explored via computationally expensive stochastic assembly models, giving a fitness landscape in terms of the assembly efficiency. Using latest machine-learning techniques by establishing a neural network, we show that the intensive computation can be short-circuited in a matter of minutes to astounding accuracy.

scholarly journals Can precision medicine advance psychiatry?

2020 ◽  
pp. 1-6
Author(s):  
Dónal Roche ◽  
Vincent Russell
Keyword(s):  
Precision Medicine ◽  
Therapeutic Interventions ◽  
Machine Learning Techniques ◽  
Major Depressive ◽  
New Approach ◽  
Considerable Potential ◽  
Traditional Research ◽  
Learning Techniques ◽  
Insight Into

Precision medicine is a new approach that considers differences in genes, environment, and lifestyle in an attempt to tailor treatments for individual patients. Psychiatry, as a discipline, has historically relied on clinical judgement and phenomenology-based diagnostic guidelines and has yet to take full advantage. This editorial provides an insight into the expanding role of precision medicine in psychiatry, both in research and clinical practice. It discusses the application of genetics and subgroup stratification in increasing response rates to therapeutic interventions, mainly focusing on major depressive disorder and schizophrenia. It presents an overview of machine learning techniques and how they are being integrated with traditional research methods within the field. In the context of these developments, while emphasizing the considerable potential for moving toward precision psychiatry, we also acknowledge the inherent challenges.

scholarly journals Recent Advancements in Various Machine Learning Techniques

2019 ◽  
Vol 8 (9S4) ◽  
pp. 10-19
Keyword(s):  
Machine Learning ◽  
House Price ◽  
Parameter Tuning ◽  
Primary Objective ◽  
Machine Learning Techniques ◽  
Local Optima ◽  
Price Prediction ◽  
Learning Techniques ◽  
Poor Convergence ◽  
Metaheuristic Techniques

The field of machine learning is witnessing its golden era as it turnout to be the leader in this field of artificial intelligence. This paper presents a comprehensive study of recently developed machine learning techniques especially for house price prediction. Due to the lack of sufficient knowledge required to train the machine learning models, existing techniques usually use various attributes and assign constant values to these attributes. Unsuitable assignment to these attributes does not provide desired results. The primary objective of this review paper is to provide a structured outline of some well-known machine learning techniques. This paper also focuses on the methods which can assign optimal values to the existing techniques. The review has revealed that the meta-heuristic techniques can attain the optimistic parameters for the machine learning techniques. However, metaheuristic techniques still suffer from the poor convergence speed and stuck in local optima kind of issues. Finally, this paper describes the various issues and challenges of image machine learning techniques, which are required to be further studied. The various challenges with existing machine learning techniques are as: parameter tuning, ensembling, over/under-fitting, etc.

Sign in / Sign up

Export Citation Format

Share Document