Pierre Parrend

A hybrid optimization tool for active magnetic regenerator

Abstract

Active Magnetic Regenerator (AMR) refrigeration is an innovate technology, which can reduce energy consumption and the depletion of the ozone layer. However, to develop a commercially applicable design of the AMR model is still an issue, because of the difficulty to find a configuration of the AMR parameters, which are suitable for various applications needs. In this work, we focus on the optimization method for finding a common parameters of the AMR model in two application modes: a magnetic refrigeration system and a thermo-magnetic generator. This paper proposes a robust optimisation tool, which ensures the scalability with respect to the number of objectives and allows to easily set up different optimisation experiments. A tool validation is presented. It is expected that this tool can help to make a qualitative jump in the development of AMR refrigeration.

Continue reading

Anomaly detection on static and dynamic graphs using graph convolutional neural networks

Abstract

Anomalies represent rare observations that vary significantly from others. Anomaly detection intended to discover these rare observations has the power to prevent detrimental events, such as financial fraud, network intrusion, and social spam. However, conventional anomaly detection methods cannot handle this problem well because of the complexity of graph data (e.g., irregular structures, relational dependencies, node/edge types/attributes/directions/multiplicities/weights, large scale, etc.) [1]. Thanks to the rise of deep learning in solving these limitations, graph anomaly detection with deep learning has obtained an increasing attention from many scientists recently. However, while deep learning can capture unseen patterns of multi-dimensional Euclidean data, there is a huge number of applications where data are represented in the form of graphs. Graphs have been used to represent the structural relational information, which raises the graph anomaly detection problem - identifying anomalous graph objects (i.e., vertex, edges, sub-graphs, and change detection). These graphs can be constructed as a static graph, or a dynamic graph based on the availability of timestamp. Recent years have observed a huge efforts on static graphs, among which Graph Convolutional Network (GCN) has appeared as a useful class of models. A challenge today is to detect anomalies with dynamic structures. In this chapter, we aim at providing methods used for detecting anomalies in static and dynamic graphs using graph analysis, graph embedding, and graph convolutional neural networks. For static graphs we categorize these methods according to plain and attribute static graphs. For dynamic graphs we categorize existing methods according to the type of anomalies that they can detect. Moreover, we focus on the challenges in this research area and discuss the strengths and weaknesses of various methods in each category. Finally, we provide open challenges for graph anomaly detection using graph convolutional neural networks on dynamic graphs.

Continue reading

GenIDA, une base de données participative internationale permettant de mieux connaître l’histoire naturelle et les comorbidités des formes génétiques de troubles neurodéveloppementaux

Abstract

Continue reading

Accurate and interpretable representations of environments with anticipatory learning classifier systems

Abstract

Anticipatory Learning Classifier Systems (ALCS) are rule- based machine learning algorithms that can simultaneously develop a complete representation of their environment and a decision policy based on this representation to solve their learning tasks. This paper intro- duces BEACS (Behavioral Enhanced Anticipatory Classifier System) in order to handle non-deterministic partially observable environments and to allow users to better understand the environmental representations issued by the system. BEACS is an ALCS that enhances and merges Probability-Enhanced Predictions and Behavioral Sequences approaches used in ALCS to handle such environments. The Probability-Enhanced Predictions consist in enabling the anticipation of several states, while the Behavioral Sequences permits the construction of sequences of ac- tions. The capabilities of BEACS have been studied on a thorough bench- mark of 23 mazes and the results show that BEACS can handle different kinds of non-determinism in partially observable environments, while describing completely and more accurately such environments. BEACS thus provides explanatory insights about created decision polices and environmental representations.

Continue reading

One-class ant-miner: Selection of majority class rules for binary rule-based classification

Abstract

In recent years, high-performance models have been introduced based on deep learning; however, these models do not have high interpretability to complement their high efficiency. Rule-based classifiers can be used to obtain explainable artificial intelligence. Rule-based classifiers use a labeled dataset to extract rules that express the relationships between inputs and expected outputs. Although many evolutionary and non-evolutionary algorithms have developed to solve this problem, we hypothesize that rule-based evolutionary algorithms such as the AntMiner family can provide good approximate solutions to problems that cannot be addressed efficiently using other techniques. This study proposes a novel supervised rule-based classifier for binary classification tasks and evaluates the extent to which algorithms in the AntMiner family can address this problem. First, we describe different versions of AntMiner. We then introduce the one-class AntMiner (OCAntMiner) algorithm, which can work with different imbalance ratios. Next, we evaluate these algorithms using specific synthetic datasets based on the AUPRC, AUROC, and MCC evaluation metrics and rank them based on these metrics. The results demonstrate that the OCAntMiner algorithm performs better than other versions of AntMiner in terms of the specified metrics.

Continue reading

Evaluation of anomaly detection for cybersecurity using inductive node embedding with convolutional graph neural networks

Abstract

In the face of continuous cyberattacks, many scientists have proposed machine learning-based network anomaly detection methods. While deep learning effectively captures unseen patterns of Euclidean data, there is a huge number of applications where data are described in the form of graphs. Graph analysis have improved detecting anomalies in non-Euclidean domains, but it suffered from high computational cost. Graph embeddings have solved this problem by converting each node in the network into low dimensional representation, but it lacks the ability to generalize to unseen nodes. Graph convolution neural network methods solve this problem through inductive node embedding (inductive GNN). Inductive GNN shows better performance in detecting anomalies with less complexity than graph analysis and graph embedding methods.

Continue reading

VizNN: Visual data augmentation with convolutional neural networks for cybersecurity investigation

Abstract

One of the key challenges of Security Operating Centers (SOCs) is to provide rich information to the security analyst to ease the investigation phase in front of a cyberattack. This requires the combination of supervision with detection capabilities. Supervision enables the security analysts to gain an overview on the security state of the information system under protection. Detection uses advanced algorithms to extract suspicious events from the huge amount of traces produced by the system. To enable coupling an efficient supervision with performance detection, the use of visualisation-based analysis is a appealing approach, which into the bargain provides an elegant solution for data augmentation and thus improved detection performance. We propose VizNN, a Convolutional Neural Networks for analysing trace features through their graphical representation. VizNN enables to gain a visual overview of the traces of interests, and Convolutional Neural Networks leverage a scalability capability. An evaluation of the proposed scheme is performed against reference classifiers for detecting attacks, XGBoost and Random Forests

Continue reading

Graph-based intelligent cyber threat detection system

Abstract

In the wake of the generalised spread of machine learning approaches, attackers are actively considering those approaches to avoid being detected. Classification models for attack detection are foremost composed of feature-driven algorithms. Thus, primary features which are individual dimension in the original attributes of data in the input space are a prime target to compromise an AI-driven model. Additionally, adversarial examples have shown that an attacker does not need to have knowledge of detection criteria to compromise a detection model, even in the case of a black box model. Attacks behavioural changes cause features from attacks datapoints to be altered and detection performances to drop. Thus, robust features must be engineered to prevent models to be compromised in such manner. Graph-based feature engineering has recently shown promising results considering robust threat detection. We offer an overview on methods for graph-based features extraction and explain why they are relevant to robust feature engineering for threat detection purposes. We detail what we think are properties for feature space to be sustainable and efficient for their prolonged exploitation in security operating centres. Specifically, we provide key criteria for the robustness of a feature space for attack detection. Finally, we summarize the characteristics for time robust feature selection, identify current limitations specific to the distinctive type of graph-based approaches in the purposes of threat detection in large internet networks.

Continue reading