A Lipschitz - Shapley Explainable Defense Methodology Against Adversarial Attacks

Demertzis, Konstantinos; Iliadis, Lazaros; Kikiras, Panayotis

doi:10.1007/978-3-030-79157-5_18

Cited by 8 publications

(8 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, it is common to use the probability logarithm l n (L n ) instead of the probability itself, as this practice leads to simpler calculations. e result, of course, remains the same as the logarithmic function is genuinely increasing [25,33,34].…”

Section: The Proposed Hybrid Deep Learning Htmmentioning

confidence: 73%

“…When one or more cells in column i are in a predicted state during the t-th iteration of the algorithm, this can be interpreted as a system prediction that the next input element might trigger column i. erefore, if the activation of this column is observed during the repetition t + 1, then the prediction of the system is considered successful with the algorithm proceeding with the activation of the cells of the column which are in the predicted state. If, on the other hand, the column is not activated, then this means that the prediction turned out to be wrong, resulting in a reduction in the permanence values of the conclusions that contributed to its execution as a kind of "punishment" [17,24,25].…”

Section: The Proposed Hybrid Deep Learning Htmmentioning

confidence: 99%

See 1 more Smart Citation

DDoS Attack Detection by Hybrid Deep Learning Methodologies

Chen

2022

Security and Communication Networks

View full text Add to dashboard Cite

A Distributed Denial of Service (DDoS) attack occurs when large amounts of traffic from hundreds, thousands, or even millions of other computers are routed to a network or server to crash the system and disrupt its function. These attacks are commonly used to shut down websites or applications temporarily. Such problems often need to be addressed with models that can manage the time information contained in network traffic flows. In this work, we apply a Hybrid Deep Learning method to detect malicious web traffic in the form of DDoS attacks, controlling the web flow of information reaching a server, using any dependencies between the different elements of a data stream. An original and cutting-edge Hierarchical Temporal Memory (HTM) hybrid model has been proposed. The operation of this model is predicated primarily on the portion of the cerebral cortex known as the neocortex. The neocortex is in charge of various fundamental brain functions, including the perception of senses, the comprehension of language, and the control of movement. For the hybrid implementation to be capable of encoding time sequences that incorporate incoming data, a Long Short-Term Memory (LSTM) shell is added.

show abstract

Section: The Proposed Hybrid Deep Learning Htmmentioning

confidence: 73%

Section: The Proposed Hybrid Deep Learning Htmmentioning

confidence: 99%

DDoS Attack Detection by Hybrid Deep Learning Methodologies

Chen

2022

Security and Communication Networks

View full text Add to dashboard Cite

show abstract

“…To evaluate and validate the final model capability resulting from applying the proposed methodology, we use the [29,30,31] constant, which allows us to study the behavior of the scattering transformation when a set of similar inputs is introduced as input. This transformation can approximate the operation of a simple neural network architecture by allowing the study of how neural networks succeed in solving difficult problems in which multiscale feature extraction is required.…”

Section: The Solution Procedures Based On the Nasad Approachmentioning

confidence: 99%

“…So, for each of the three kinds of layers, we have: 1) Let layer 𝑘 is a convolution layer. As we express the inputs as one-dimensional vectors, the convolution with a two-dimensional kernel 𝜓 𝑖𝑗𝑘 , which connects the i channel of the output to the 𝑗 channel of the input, is implemented by multiplying the input vector by a matrix 𝐴 𝑖𝑗𝑘 generated by the original kernel, such that [30,34]:…”

Section: The Solution Procedures Based On the Nasad Approachmentioning

confidence: 99%

The Dynamic Problem of an Elastically Supported Beam Analysis Using Neural Architecture Search With Automatic Differentiation

Demertzis,

Morfidis,

Kostinakis

et al. 2023

Proceedings of the 8th International Conference on Computational Methods in Structural Dynamics and Earthquake Engineering (COM

View full text Add to dashboard Cite

The equations of motion for linear structures considered as continuous systems are generally non-homogeneous linear partial differential equations. Their solution is functions of time and spatial variables. The traditional analysis approach involves solving the eigenvalue problem. This yields the eigenvalues and eigenvectors, which can be used to construct the general solution of the differential equations. This approach has been widely used in mechanics and has proven effective in solving many engineering problems. In recent years, machine learning methods, particularly neural networks, have also been used to solve differential equations. These methods can learn the equation's solution directly from data without the need for explicit analytical expressions. This makes them particularly useful for complex problems that may need analytical solutions or for problems where obtaining analytical solutions takes time and effort. This paper compares the performance of these two approaches for a specific problem of the vibration of the Euler-Bernoulli beam on a Winkler-type elastic foundation subjected to a moving load. The machine learning approach learns the solution directly from data generated by solving the differential equation using a Neural Architecture Search (NAS) with Automatic Differentiation (AD) method (NASAD). The paper provides insights into the relative strengths and weaknesses of each method and highlights the potential of machine learning to solve complex problems with high accuracy and low computational sources.

show abstract

“…It is a severe concern in the reliability and security of artificial intelligence technologies. The issue arises because learning techniques are intended for use in stable situations where training and test data are generated from the same, possibly unknown distribution [ 3 ]. A trained neural network, for example, represents a significant decision limit corresponding to a standard class.…”

Section: Introductionmentioning

confidence: 99%

Leakage Prediction in Machine Learning Models When Using Data from Sports Wearable Sensors

Qi-zheng

2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

One of the major problems in machine learning is data leakage, which can be directly related to adversarial type attacks, raising serious concerns about the validity and reliability of artificial intelligence. Data leakage occurs when the independent variables used to teach the machine learning algorithm include either the dependent variable itself or a variable that contains clear information that the model is trying to predict. This data leakage results in unreliable and poor predictive results after the development and use of the model. It prevents the model from generalizing, which is required in a machine learning problem and thus causes false assumptions about its performance. To have a solid and generalized forecasting model, which will be able to produce remarkable forecasting results, we must pay great attention to detecting and preventing data leakage. This study presents an innovative system of leakage prediction in machine learning models, which is based on Bayesian inference to produce a thorough approach to calculating the reverse probability of unseen variables in order to make statistical conclusions about the relevant correlated variables and to calculate accordingly a lower limit on the marginal likelihood of the observed variables being derived from some coupling method. The main notion is that a higher marginal probability for a set of variables suggests a better fit of the data and thus a greater likelihood of a data leak in the model. The methodology is evaluated in a specialized dataset derived from sports wearable sensors.

show abstract

A Lipschitz - Shapley Explainable Defense Methodology Against Adversarial Attacks

Cited by 8 publications

References 24 publications

DDoS Attack Detection by Hybrid Deep Learning Methodologies

DDoS Attack Detection by Hybrid Deep Learning Methodologies

The Dynamic Problem of an Elastically Supported Beam Analysis Using Neural Architecture Search With Automatic Differentiation

Leakage Prediction in Machine Learning Models When Using Data from Sports Wearable Sensors

Contact Info

Product

Resources

About