arXiv subjects

Luis Garcia

Publications and source records attributed to Luis Garcia.

28 records · Page 2Linked to original sources

arXiv2021

Using DeepProbLog to perform Complex Event Processing on an Audio Stream

In this paper, we present an approach to Complex Event Processing (CEP) that is based on DeepProbLog. This approach has the following objectives: (i) allowing the use of subsymbolic data as an input, (ii) retaining the flexibility and modularity on the definitions of complex event rules, (iii) allowing the system to be trained in an end-to-end manner and (iv) being robust against noisily labelled data. Our approach makes use of DeepProbLog to create a neuro-symbolic architecture that combines a neural network to process the subsymbolic data with a probabilistic logic layer to allow the user to define the rules for the complex events. We demonstrate that our approach is capable of detecting complex events from an audio stream. We also demonstrate that our approach is capable of training even with a dataset that has a moderate proportion of noisy data.

cs.SD↗

arXiv2021

I Always Feel Like Somebody's Sensing Me! A Framework to Detect, Identify, and Localize Clandestine Wireless Sensors

The increasing ubiquity of low-cost wireless sensors has enabled users to easily deploy systems to remotely monitor and control their environments. However, this raises privacy concerns for third-party occupants, such as a hotel room guest who may be unaware of deployed clandestine sensors. Previous methods focused on specific modalities such as detecting cameras but do not provide a generalized and comprehensive method to capture arbitrary sensors which may be "spying" on a user. In this work, we propose SnoopDog, a framework to not only detect common Wi-Fi-based wireless sensors that are actively monitoring a user, but also classify and localize each device. SnoopDog works by establishing causality between patterns in observable wireless traffic and a trusted sensor in the same space, e.g., an inertial measurement unit (IMU) that captures a user's movement. Once causality is established, SnoopDog performs packet inspection to inform the user about the monitoring device. Finally, SnoopDog localizes the clandestine device in a 2D plane using a novel trial-based localization technique. We evaluated SnoopDog across several devices and various modalities and were able to detect causality for snooping devices 95.2% of the time and localize devices to a sufficiently reduced sub-space.

cs.CR↗

arXiv2020

An Experimentation Platform for Explainable Coalition Situational Understanding

We present an experimentation platform for coalition situational understanding research that highlights capabilities in explainable artificial intelligence/machine learning (AI/ML) and integration of symbolic and subsymbolic AI/ML approaches for event processing. The Situational Understanding Explorer (SUE) platform is designed to be lightweight, to easily facilitate experiments and demonstrations, and open. We discuss our requirements to support coalition multi-domain operations with emphasis on asset interoperability and ad hoc human-machine teaming in a dense urban terrain setting. We describe the interface functionality and give examples of SUE applied to coalition situational understanding tasks.

cs.AI↗

arXiv2020

A Hybrid Neuro-Symbolic Approach for Complex Event Processing

Training a model to detect patterns of interrelated events that form situations of interest can be a complex problem: such situations tend to be uncommon, and only sparse data is available. We propose a hybrid neuro-symbolic architecture based on Event Calculus that can perform Complex Event Processing (CEP). It leverages both a neural network to interpret inputs and logical rules that express the pattern of the complex event. Our approach is capable of training with much fewer labelled data than a pure neural network approach, and to learn to classify individual events even when training in an end-to-end manner. We demonstrate this comparing our approach against a pure neural network approach on a dataset based on Urban Sounds 8K.

cs.AI↗

arXiv2019

HyPLC: Hybrid Programmable Logic Controller Program Translation for Verification

Programmable Logic Controllers (PLCs) provide a prominent choice of implementation platform for safety-critical industrial control systems. Formal verification provides ways of establishing correctness guarantees, which can be quite important for such safety-critical applications. But since PLC code does not include an analytic model of the system plant, their verification is limited to discrete properties. In this paper, we, thus, start the other way around with hybrid programs that include continuous plant models in addition to discrete control algorithms. Even deep correctness properties of hybrid programs can be formally verified in the theorem prover KeYmaera X that implements differential dynamic logic, dL, for hybrid programs. After verifying the hybrid program, we now present an approach for translating hybrid programs into PLC code. The new tool, HyPLC, implements this translation of discrete control code of verified hybrid program models to PLC controller code and, vice versa, the translation of existing PLC code into the discrete control actions for a hybrid program given an additional input of the continuous dynamics of the system to be verified. This approach allows for the generation of real controller code while preserving, by compilation, the correctness of a valid and verified hybrid program. PLCs are common cyber-physical interfaces for safety-critical industrial control applications, and HyPLC serves as a pragmatic tool for bridging formal verification of complex cyber-physical systems at the algorithmic level of hybrid programs with the execution layer of concrete PLC implementations.

cs.PL↗

arXiv2018

Control Behavior Integrity for Distributed Cyber-Physical Systems

Cyber-physical control systems, such as industrial control systems (ICS), are increasingly targeted by cyberattacks. Such attacks can potentially cause tremendous damage, affect critical infrastructure or even jeopardize human life when the system does not behave as intended. Cyberattacks, however, are not new and decades of security research have developed plenty of solutions to thwart them. Unfortunately, many of these solutions cannot be easily applied to safety-critical cyber-physical systems. Further, the attack surface of ICS is quite different from what can be commonly assumed in classical IT systems. We present Scadman, a system with the goal to preserve the Control Behavior Integrity (CBI) of distributed cyber-physical systems. By observing the system-wide behavior, the correctness of individual controllers in the system can be verified. This allows Scadman to detect a wide range of attacks against controllers, like programmable logic controller (PLCs), including malware attacks, code-reuse and data-only attacks. We implemented and evaluated Scadman based on a real-world water treatment testbed for research and training on ICS security. Our results show that we can detect a wide range of attacks--including attacks that have previously been undetectable by typical state estimation techniques--while causing no false-positive warning for nominal threshold values.

cs.CR↗

arXiv2009

Zonal flows and long-distance correlations during the formation of the edge shear layer in the TJ-II stellarator

A theoretical interpretation is given for the observed long-distance correlations in potential fluctuations in TJ-II. The value of the correlation increases above the critical point of the transition for the emergence of the plasma edge shear flow layer. Mean (i.e. surface averaged, zero-frequency) sheared flows cannot account for the experimental results. A model consisting of four envelope equations for the fluctuation level, the mean flow shear, the zonal flow amplitude shear, and the averaged pressure gradient is proposed. It is shown that the presence of zonal flows is essential to reproduce the main features of the experimental observations.

physics.plasm-ph↗

arXiv2008

Topological characterization of flow structures in resistive pressure-gradient-driven turbulence

Visualization of turbulent flows is a powerful tool to help understand the turbulence dynamics and induced transport. However, it does not provide a quantitative description of the observed structures. In this paper, an approach to characterize quantitatively the topology of the flows is given. The technique, which can be applied to any type of turbulence dynamics, is illustrated through the example of resistive ballooning instabilities.

physics.plasm-ph↗

arXiv2008

Pseudochaotic poloidal transport in the laminar regime of the resistive ballooning instabilities

In toroidal geometry, and prior to the establishment of a fully developed turbulent state, the so-called topological instability of the pressure-gradient-driven turbulence is observed. In this intermediate state, a narrow spectral band of modes dominates the dynamics, giving rise to the formation of iso-surfaces of electric potential with a complicated topology. Since E x B advection of tracer particles takes place along these iso-surfaces, their topological complexity affects the characteristic features of radial and poloidal transport dramatically. In particular, they both become strongly non-diffusive and non-Gaussian. Since radial transport determines the system confinement properties and poloidal transport controls the equilibration dynamics (on any magnetic surface), the development of non-diffusive models in both directions is thus of physical interest. In previous work, a fractional model to describe radial transport was constructed by the authors. In this contribution, recent results on periodic fractional models are exploited for the construction of an effective model of poloidal transport. Numerical computations using a three-dimensional reduced magnetohydrodynamic set of equations are compared with analytical solutions of the fractional periodic model. It is shown that the aforementioned analytical solutions accurately describe poloidal transport, which turns out to be superdiffusive with index $α=1$.

physics.plasm-ph↗

arXiv2001

Classification of Finite Dynamical Systems

This paper is motivated by the theory of sequential dynamical systems, developed as a basis for a mathematical theory of computer simulation. It contains a classification of finite dynamical systems on binary strings, which are obtained by composing functions defined on the coordinates. The classification is in terms of the dependency relations among the coordinate functions. It suggests a natural notion of the linearization of a system. Furthermore, it contains a sharp upper bound on the number of systems in terms of the dependencies among the coordinate functions. This upper bound generalizes an upper bound for sequential dynamical systems.

math.DS↗