Search for: All records

Graph Autoencoder-Based Detection of Unseen False Data Injection Attacks in Smart Grids Abdulrahman Takiddin, Muhammad Ismail, Rachad Atat, Katherine R. Davis, Erchin Serpedin Pages 234-244 

Discovering causal knowledge is an important aspect of much scientific research and such findings are often recorded in scholarly articles. Automatically identifying such knowledge from article text can be a useful tool and can act as an impetus for further research on those topics. Numerous applications, including building a causal knowledge graph, making pipelines for root cause analysis, discovering opportunities for drug discovery, and overall, a scalable building block towards turning large pieces of text into organized information can be built following such an approach. However, it requires robust methods to identify and aggregate causal knowledge from a large set of articles. The main challenge in designing new methods is the absence of a large labeled dataset. As a result, existing methods trained on existing datasets with limited size and variations in linguistic pattern, are unable to generalize well on unseen text. In this paper, we explore multiple unsupervised approaches, including a reinforcement learning-based model that learns to identify causal sentences from a small set of labeled sentences. We describe and discuss in detail our experiments for each approach to further encourage exploration of methods that can be re-utilized for different tasks as well, as opposed to simply exploring a supervised learning process which although superior in performance lacks the versatility to be re-purposed for slightly different tasks. We evaluate our methods on a custom-created dataset and show unique techniques to extract cause-effect relationships from the English language. 

Quantum noise is seen by many researchers as a problem to be resolved. Current solutions increase quantum computing system costs significantly by requiring numerous hardware qubits to represent a logical qubit to average the noise away. However, despite its deleterious effects on system performance and the increased costs it creates, it may have some potential uses. This paper evaluates those. Specifically, it considers how quantum noise could be used to support the fuzzing cybersecurity and testing technique and AI techniques such as certain swarm artificial intelligence algorithms. Fuzzing is used to identify vulnerabilities in software by generating massive amounts of input cases for a program. Quantum noise provides an effective built-in fuzzing capability that is centered around the actual answer to a computation. These same phenomena, of clustered and centered fuzz-noise around the answer of an operation, could be similarly useful to AI techniques that can make effective use of lots of point values for optimization. Effectively, by concurrently considering the ‘multiverse’ of possible results to an operation, created by compounding noise, more beneficial solutions that are proximal to the actual result of an operation can be identified via testing quantum noise points with an effectiveness algorithm. Both of these potential uses for quantum noise are considered herein. 

This Quantum Machine Learning Classifier (QMLC) uses the mathematics of quantum computing in a deep neural network to find and classify the specific flower type of the three different iris flower species: Versicolor, Setosa and Virginica, utilizing the SciKit-Learn dataset ``Iris.'' In that dataset, there are four characteristic features of each iris type: petal length, petal width, sepal length, and sepal width. The quantum computing machine learning classifier out-performed the classical deep learning neural network methods. Significant is that this classifier trained in fewer epochs. 

This research compares and contrasts two commonly available quantum computing platforms available today to academic researchers: the IBM Q-Experience and the University of Maryland's IonQ. Hands-on testing utilized the implementation of a simple two qubit circuit and tested the Pauli X, Y, and Z single-qubit gates as well as the CNOT 2+ qubit gate and compared the results, as well as the user experience. The user experience and the interface must be straightforward to help the user's understanding when planning quantum computing training for new knowledge workers in this exciting new field. Additionally, we demonstrate how a quantum computer's results, when the output is read in the classical computer, loses some of its information, since the quantum computer is operating in more dimensions than the classical computer can interpret. This is demonstrated with the ZX and XZ gates which appear to give the same result; however, using the mathematics of matrix notation, the phase difference between the two answers is revealed in their vectors, which are 180 degrees apart. 

Scientific Machine Learning (SciML) is a new multidisciplinary methodology that combines the data-driven machine learning models and the principle-based computational models to improve the simulations of scientific phenomenon and uncover new scientific rules from existing measurements. This article reveals the experience of using the SciML method to discover the nonlinear dynamics that may be hard to model or be unknown in the real-world scenario. The SciML method solves the traditional principle-based differential equations by integrating a neural network to accurately model the nonlinear dynamics while respecting the scientific constraints and principles. The paper discusses the latest SciML models and apply them to the oscillator simulations and experiment. Besides better capacity to simulate, and match with the observation, the results also demonstrate a successful discovery of the hidden physics in the pendulum dynamics using SciML.