Search

Quantum machine learning has gained attention for its potential to address computational challenges. However, whether those algorithms can effectively solve practical problems and outperform their classical counterparts, especially on current quantum hardware, remains a critical question. In this work, we propose a novel quantum machine learning method, called Quantum Support Vector Data Description (QSVDD), for practical image anomaly detection, which aims to achieve both parameter efficiency and superior accuracy compared to classical models. Emulation results indicate that QSVDD demonstrates favourable recognition capabilities compared to classical baselines, achieving an average accuracy of over 90% on benchmarks with significantly fewer trainable parameters. Theoretical analysis confirms that QSVDD has a comparable expressivity to classical counterparts while requiring only a fraction of the parameters. Furthermore, we demonstrate the first implementation of a quantum anomaly detection method for general image datasets on a superconducting quantum processor. Specifically, we achieve an accuracy of over 80% with only 16 parameters on the device, providing initial evidence of QSVDD's practical viability in the noisy intermediate-scale quantum era and highlighting its significant reduction in parameter requirements., Comment: 30 pages, 13 figures

We construct a thermodynamic limit for the grand canonical Bose gas (in its Feynman-Kac representation) with superstable interaction. Although we do not prove the presence of infinite cycles, our infinite volume model is naturally a distribution over configurations of finite loops and interlacements. We prove the limiting process to be solution of DLR equations. We will work within the framework of Dirichlet and periodic boundary conditions, for any inverse temperature $\beta$ > 0, chemical potential $\mu$ $\in$ R and dimension d$\ge$1.

Since the onset of LLMs, translating natural language queries to structured SQL commands is assuming increasing. Unlike the previous reviews, this survey provides a comprehensive study of the evolution of LLM-based text-to-SQL systems, from early rule-based models to advanced LLM approaches, and how LLMs impacted this field. We discuss benchmarks, evaluation methods and evaluation metrics. Also, we uniquely study the role of integration of knowledge graphs for better contextual accuracy and schema linking in these systems. The current techniques fall into two categories: in-context learning of corpus and fine-tuning, which then leads to approaches such as zero-shot, few-shot learning from the end, and data augmentation. Finally, we highlight key challenges such as computational efficiency, model robustness, and data privacy with perspectives toward their development and improvements in potential areas for future of LLM-based text-to-SQL system., Comment: 12 pages, 5 figures, 3 tables

Image compression for machine and human vision (ICMH) has gained increasing attention in recent years. Existing ICMH methods are limited by high training and storage overheads due to heavy design of task-specific networks. To address this issue, in this paper, we develop a novel lightweight adapter-based tuning framework for ICMH, named Adapt-ICMH, that better balances task performance and bitrates with reduced overheads. We propose a spatial-frequency modulation adapter (SFMA) that simultaneously eliminates non-semantic redundancy with a spatial modulation adapter, and enhances task-relevant frequency components and suppresses task-irrelevant frequency components with a frequency modulation adapter. The proposed adapter is plug-and-play and compatible with almost all existing learned image compression models without compromising the performance of pre-trained models. Experiments demonstrate that Adapt-ICMH consistently outperforms existing ICMH frameworks on various machine vision tasks with fewer fine-tuned parameters and reduced computational complexity. Code will be released at https://github.com/qingshi9974/ECCV2024-AdpatICMH ., Comment: Accepted by ECCV2024, project: https://github.com/qingshi9974/ECCV2024-AdpatICMH

We study a Gaussian measure with parameter $q\in(0,1)$ on the dual of the unitary group of size $N$: we prove that a random highest weight under this measure is the coupling of two independent $q$-uniform random partitions $\alpha,\beta$ and a random highest weight of $\mathrm{U}(1)$. We prove deviation inequalities for the $q$-uniform measure, and use them to show that the coupling of random partitions under the Gaussian measure vanishes in the limit $N\to\infty$. We also prove that the partition function of this measure admits an asymptotic expansion in powers of $1/N$, and that this expansion is topological, in the sense that its coefficients are related to the enumeration of ramified coverings of elliptic curves. It provides a rigorous proof of the gauge/string duality for the Yang-Mills theory on a 2D torus with gauge group $\mathrm{U}(N),$ advocated by Gross and Taylor \cite{GT,GT2}.

Kernel herding belongs to a family of deterministic quadratures that seek to minimize the worst-case integration error over a reproducing kernel Hilbert space (RKHS). In spite of strong experimental support, it has revealed difficult to prove that this worst-case error decreases at a faster rate than the standard square root of the number of quadrature nodes, at least in the usual case where the RKHS is infinite-dimensional. In this theoretical paper, we study a joint probability distribution over quadrature nodes, whose support tends to minimize the same worst-case error as kernel herding. We prove that it does outperform i.i.d. Monte Carlo, in the sense of coming with a tighter concentration inequality on the worst-case integration error. While not improving the rate yet, this demonstrates that the mathematical tools of the study of Gibbs measures can help understand to what extent kernel herding and its variants improve on computationally cheaper methods. Moreover, we provide early experimental evidence that a faster rate of convergence, though not worst-case, is likely., Comment: 16 pages, 2 figures. Comments are welcome

The development of various sensing technologies is improving measurements of stress and the well-being of individuals. Although progress has been made with single signal modalities like wearables and facial emotion recognition, integrating multiple modalities provides a more comprehensive understanding of stress, given that stress manifests differently across different people. Multi-modal learning aims to capitalize on the strength of each modality rather than relying on a single signal. Given the complexity of processing and integrating high-dimensional data from limited subjects, more research is needed. Numerous research efforts have been focused on fusing stress and emotion signals at an early stage, e.g., feature-level fusion using basic machine learning methods and 1D-CNN Methods. This paper proposes a multi-modal learning approach for stress detection that integrates facial landmarks and biometric signals. We test this multi-modal integration with various early-fusion and late-fusion techniques to integrate the 1D-CNN model from biometric signals and 2-D CNN using facial landmarks. We evaluate these architectures using a rigorous test of models' generalizability using the leave-one-subject-out mechanism, i.e., all samples related to a single subject are left out to train the model. Our findings show that late-fusion achieved 94.39\% accuracy, and early-fusion surpassed it with a 98.38\% accuracy rate. This research contributes valuable insights into enhancing stress detection through a multi-modal approach. The proposed research offers important knowledge in improving stress detection using a multi-modal approach., Comment: 16 pages, 8 figures

Ecosystems with a large number of species are often modelled as Lotka-Volterra dynamical systems built around a large random interaction matrix. Under some known conditions, a global equilibrium exists and is unique. In this article, we rigorously study its statistical properties in the large dimensional regime. Such an equilibrium vector is known to be the solution of a so-called Linear Complementarity Problem (LCP). We describe its statistical properties by designing an Approximate Message Passing (AMP) algorithm, a technique that has recently aroused an intense research effort in the fields of statistical physics, Machine Learning, or communication theory. Interaction matrices taken from the Gaussian Orthogonal Ensemble, or following a Wishart distribution are considered. Beyond these models, the AMP approach developed in this article has the potential to describe the statistical properties of equilibria associated to more involved interaction matrix models.

While convolutional neural networks (CNNs) demonstrate outstanding performance on computer vision tasks, their computational costs remain high. Several techniques are used to reduce these costs, like reducing channel count, and using separable and depthwise separable convolutions. This paper reduces computational costs by introducing a novel architecture, axial CNNs, which replaces spatial 2D convolution operations with two consecutive depthwise separable 1D operations. The axial CNNs are predicated on the assumption that the dataset supports approximately separable convolution operations with little or no loss of training accuracy. Deep axial separable CNNs still suffer from gradient problems when training deep networks. We modify the construction of axial separable CNNs with residual connections to improve the performance of deep axial architectures and introduce our final novel architecture namely residual axial networks (RANs). Extensive benchmark evaluation shows that RANs achieve at least 1% higher performance with about 77%, 86%, 75%, and 34% fewer parameters and about 75%, 80%, 67%, and 26% fewer flops than ResNets, wide ResNets, MobileNets, and SqueezeNexts on CIFAR benchmarks, SVHN, and Tiny ImageNet image classification datasets. Moreover, our proposed RANs improve deep recursive residual networks performance with 94% fewer parameters on the image super-resolution dataset.

Over the past decade, deep hypercomplex-inspired networks have enhanced feature extraction for image classification by enabling weight sharing across input channels. Recent works make it possible to improve representational capabilities by using hypercomplex-inspired networks which consume high computational costs. This paper reduces this cost by factorizing a quaternion 2D convolutional module into two consecutive vectormap 1D convolutional modules. Also, we use 5D parameterized hypercomplex multiplication based fully connected layers. Incorporating both yields our proposed hypercomplex network, a novel architecture that can be assembled to construct deep axial-hypercomplex networks (DANs) for image classifications. We conduct experiments on CIFAR benchmarks, SVHN, and Tiny ImageNet datasets and achieve better performance with fewer trainable parameters and FLOPS. Our proposed model achieves almost 2% higher performance for CIFAR and SVHN datasets, and more than 3% for the ImageNet-Tiny dataset and takes six times fewer parameters than the real-valued ResNets. Also, it shows state-of-the-art performance on CIFAR benchmarks in hypercomplex space.

Recently, many deep networks have introduced hypercomplex and related calculations into their architectures. In regard to convolutional networks for classification, these enhancements have been applied to the convolution operations in the frontend to enhance accuracy and/or reduce the parameter requirements while maintaining accuracy. Although these enhancements have been applied to the convolutional frontend, it has not been studied whether adding hypercomplex calculations improves performance when applied to the densely connected backend. This paper studies ResNet architectures and incorporates parameterized hypercomplex multiplication (PHM) into the backend of residual, quaternion, and vectormap convolutional neural networks to assess the effect. We show that PHM does improve classification accuracy performance on several image datasets, including small, low-resolution CIFAR 10/100 and large high-resolution ImageNet and ASL, and can achieve state-of-the-art accuracy for hypercomplex networks.

Long short-term memory (LSTM) is one of the robust recurrent neural network architectures for learning sequential data. However, it requires considerable computational power to learn and implement both software and hardware aspects. This paper proposed a novel LiteLSTM architecture based on reducing the LSTM computation components via the weights sharing concept to reduce the overall architecture computation cost and maintain the architecture performance. The proposed LiteLSTM can be significant for processing large data where time-consuming is crucial while hardware resources are limited, such as the security of IoT devices and medical data processing. The proposed model was evaluated and tested empirically on three different datasets from the computer vision, cybersecurity, speech emotion recognition domains. The proposed LiteLSTM has comparable accuracy to the other state-of-the-art recurrent architecture while using a smaller computation budget., Comment: Under the second reviewing round in the SN Computer Science Journal. Extended version of the LiteLSTM Architecture for Deep Recurrent Neural Networks paper that have been published in the IEEE ISCAS 2022 conference. arXiv admin note: substantial text overlap with arXiv:2201.11624

Ecosystems represent archetypal complex dynamical systems, often modelled by coupled differential equations of the form $$ \frac{d x_i}{d t} = x_i \varphi_i(x_1,\cdots, x_N)\ , $$ where $N$ represents the number of species and $x_i$, the abundance of species $i$. Among these families of coupled diffential equations, Lotka-Volterra (LV) equations $$ \frac{d x_i}{d t} = x_i ( r_i - x_i +(\Gamma \mathbf{x})_i)\ , $$ play a privileged role, as the LV model represents an acceptable trade-off between complexity and tractability. Here, $r_i$ represents the intrinsic growth of species $i$ and $\Gamma$ stands for the interaction matrix: $\Gamma_{ij}$ represents the effect of species $j$ over species $i$. For large $N$, estimating matrix $\Gamma$ is often an overwhelming task and an alternative is to draw $\Gamma$ at random, parametrizing its statistical distribution by a limited number of model features. Dealing with large random matrices, we naturally rely on Random Matrix Theory (RMT). The aim of this review article is to present an overview of the work at the junction of theoretical ecology and large random matrix theory. It is intended to an interdisciplinary audience spanning theoretical ecology, complex systems, statistical physics and mathematical biology.