A Cybersecurity Risk Analysis Framework for Systems with Artificial Intelligence Components

no code implementations • 3 Jan 2024 • Jose Manuel Camacho, Aitor Couce-Vieira, David Arroyo, David Rios Insua

The introduction of the European Union Artificial Intelligence Act, the NIST Artificial Intelligence Risk Management Framework, and related norms demands a better understanding and implementation of novel risk analysis approaches to evaluate systems with Artificial Intelligence components.

Management

Protecting Classifiers From Attacks. A Bayesian Approach

1 code implementation • 18 Apr 2020 • Victor Gallego, Roi Naveiro, Alberto Redondo, David Rios Insua, Fabrizio Ruggeri

Classification problems in security settings are usually modeled as confrontations in which an adversary tries to fool a classifier manipulating the covariates of instances to obtain a benefit.

Adversarial Machine Learning: Bayesian Perspectives

1 code implementation • 7 Mar 2020 • David Rios Insua, Roi Naveiro, Victor Gallego, Jason Poulos

Adversarial Machine Learning (AML) is emerging as a major field aimed at protecting machine learning (ML) systems against security threats: in certain scenarios there may be adversaries that actively manipulate input data to fool learning systems.

Adversarial Robustness BIG-bench Machine Learning

Protecting from Malware Obfuscation Attacks through Adversarial Risk Analysis

no code implementations • 9 Nov 2019 • Alberto Redondo, David Rios Insua

Malware constitutes a major global risk affecting millions of users each year.

Variationally Inferred Sampling Through a Refined Bound

1 code implementation • pproximateinference AABI Symposium 2019 • Victor Gallego, David Rios Insua

A framework for efficient Bayesian inference in probabilistic programs is introduced by embedding a sampler inside a variational posterior approximation.

Bayesian Inference Density Estimation +2

Variationally Inferred Sampling Through a Refined Bound for Probabilistic Programs

1 code implementation • 26 Aug 2019 • Victor Gallego, David Rios Insua

A framework to boost the efficiency of Bayesian inference in probabilistic programs is introduced by embedding a sampler inside a variational posterior approximation.

Bayesian Inference Density Estimation +2

Opponent Aware Reinforcement Learning

1 code implementation • 22 Aug 2019 • Victor Gallego, Roi Naveiro, David Rios Insua, David Gomez-Ullate Oteiza

We introduce Threatened Markov Decision Processes (TMDPs) as an extension of the classical Markov Decision Process framework for Reinforcement Learning (RL).

reinforcement-learning Reinforcement Learning (RL)

Stochastic Gradient MCMC with Repulsive Forces

2 code implementations • 30 Nov 2018 • Victor Gallego, David Rios Insua

We propose a unifying view of two different Bayesian inference algorithms, Stochastic Gradient Markov Chain Monte Carlo (SG-MCMC) and Stein Variational Gradient Descent (SVGD), leading to improved and efficient novel sampling schemes.

Bayesian Inference valid

Reinforcement Learning under Threats

1 code implementation • 5 Sep 2018 • Victor Gallego, Roi Naveiro, David Rios Insua

In several reinforcement learning (RL) scenarios, mainly in security settings, there may be adversaries trying to interfere with the reward generating process.

reinforcement-learning Reinforcement Learning (RL)