Search

This work addresses the persistent challenges of substantial training time and GPU resource requirements even when training lightweight encoder-vocoder models for Textless NLP. We reduce training steps significantly while improving performance by a) leveraging learning rate schedulers for efficient and faster convergence b) optimizing hop length and c) tuning the interpolation scale factors for better audio quality. Additionally, we explore the latent space representation for Indian languages such as Tamil and Bengali for the acoustic unit discovery and voice conversion task. Our approach leverages a quantized encoder architecture, in conjunction with a vocoder which utilizes the proposed mixture of optimized hop length, tuned interpolation scale factors and a cyclic learning rate scheduler. We obtain consistently good results across English, Tamil and Bengali datasets. The proposed method excels in capturing complex linguistic patterns, resulting in clear reconstructed audio during voice conversion with significantly reduced training time.

We propose a minimal power white box adversarial attack for Deep Learning based Automatic Modulation Classification (AMC). The proposed attack uses the Golden Ratio Search (GRS) method to find powerful attacks with minimal power. We evaluate the efficacy of the proposed method by comparing it with existing adversarial attack approaches. Additionally, we test the robustness of the proposed attack against various state-of-the-art architectures, including defense mechanisms such as adversarial training, binarization, and ensemble methods. Experimental results demonstrate that the proposed attack is powerful, requires minimal power, and can be generated in less time, significantly challenging the resilience of current AMC methods., Comment: 5 pages, 1 figure, 3 tables

In reconfigurable intelligent surface (RIS) aided systems, the joint optimization of the precoder matrix at the base station and the phase shifts of the RIS elements involves significant complexity. In this paper, we propose a complex-valued, geometry aware meta-learning neural network that maximizes the weighted sum rate in a multi-user multiple input single output system. By leveraging the complex circle geometry for phase shifts and spherical geometry for the precoder, the optimization occurs on Riemannian manifolds, leading to faster convergence. We use a complex-valued neural network for phase shifts and an Euler inspired update for the precoder network. Our approach outperforms existing neural network-based algorithms, offering higher weighted sum rates, lower power consumption, and significantly faster convergence. Specifically, it converges faster by nearly 100 epochs, with a 0.7 bps improvement in weighted sum rate and a 1.8 dBm power gain when compared with existing work.

Learning rate is a crucial parameter in training of neural networks. A properly tuned learning rate leads to faster training and higher test accuracy. In this paper, we propose a Lipschitz bandit-driven approach for tuning the learning rate of neural networks. The proposed approach is compared with the popular HyperOpt technique used extensively for hyperparameter optimization and the recently developed bandit-based algorithm BLiE. The results for multiple neural network architectures indicate that our method finds a better learning rate using a) fewer evaluations and b) lesser number of epochs per evaluation, when compared to both HyperOpt and BLiE. Thus, the proposed approach enables more efficient training of neural networks, leading to lower training time and lesser computational cost.

The performance of the standard Online Robust Principal Component Analysis (OR-PCA) technique depends on the optimum tuning of the explicit regularizers and this tuning is dataset sensitive. We aim to remove the dependency on these tuning parameters by using implicit regularization. We propose to use the implicit regularization effect of various modified gradient descents to make OR-PCA tuning free. Our method incorporates three different versions of modified gradient descent that separately but naturally encourage sparsity and low-rank structures in the data. The proposed method performs comparable or better than the tuned OR-PCA for both simulated and real-world datasets. Tuning-free ORPCA makes it more scalable for large datasets since we do not require dataset-dependent parameter tuning.

Source enumeration, the task of estimating the number of sources from the signal received by the array of antennas, is a critical problem in array signal processing. Numerous methods have been proposed to estimate the number of sources under white or colored Gaussian noise. However, their performance degrades significantly in the presence of a limited number of observations and/or a large number of sources. In this work, we propose a method based on the distribution of angles that performs well in (a) independent Gaussian, (b) spatially colored Gaussian, and (c) heavy-tailed noise, even when the number of sources is large. We support the supremacy of our algorithm over state-of-the-art methods with extensive simulation results.

A nontrivial connected graph is matching covered if each edge belongs to some perfect matching. There is extensive theory on these graphs; see Lucchesi and Murty [Perfect Matchings, Springer 2024]. Two edges are mutually dependent if every perfect matching either contains both or neither. This is an equivalence relation and induces a partition of the edges; each part is called an equivalence class. Note that $\frac{n}{2}$ is an upper bound on the size of any equivalence class. We characterize those graphs that attain this upper bound, thus fixing an error in [Discrete Mathematics, 343(8), 2020]. Sch\"onberger (1934) strengthened the famous result of Petersen by proving that every 2-connected cubic graph is matching covered. This immediately begs the question as to which 2-connected cubic graphs have the property that each edge belongs to two or more perfect matchings. An edge is solitary if it belongs to exactly one perfect matching. Note that if any member of an equivalence class is solitary then so is each member; such an equivalence class is called a solitary class. This leads us to the solitary pattern -- the sequence of sizes of solitary classes in nonincreasing order. For instance, $K_4$ has solitary pattern (2,2,2) whereas the Petersen graph has solitary pattern (). Using the theory of Lucchesi and Murty, we prove that every 3-connected cubic graph $G$ has one of the following solitary patterns: (2,2,2), (2,2,1), (2,2), (2,1,1), (2,1), (2), (1,1,1), (1,1), (1) or (); consequently, $G$ has at most six solitary edges. We also provide characterizations of 3-connected cubic graphs that have a given solitary pattern except for (1,1), (1) and (). Finally, we prove that every 2-connected cubic graph, except $\theta, K_4, \overline{C_6}$ and the bicorn, has at most $\frac{n}{2}$ solitary edges, and equality holds if and only if its solitary edges comprise a perfect matching.

Adversarial training (AT) incurs significant computational overhead, leading to growing interest in designing inherently robust architectures. We demonstrate that a carefully designed first layer of the neural network can serve as an implicit adversarial noise filter (ANF). This filter is created using a combination of large kernel size, increased convolution filters, and a maxpool operation. We show that integrating this filter as the first layer in architectures such as ResNet, VGG, and EfficientNet results in adversarially robust networks. Our approach achieves higher adversarial accuracies than existing natively robust architectures without AT and is competitive with adversarial-trained architectures across a wide range of datasets. Supporting our findings, we show that (a) the decision regions for our method have better margins, (b) the visualized loss surfaces are smoother, (c) the modified peak signal-to-noise ratio (mPSNR) values at the output of the ANF are higher, (d) high-frequency components are more attenuated, and (e) architectures incorporating ANF exhibit better denoising in Gaussian noise compared to baseline architectures. Code for all our experiments are available at \url{https://github.com/janani-suresh-97/first-line-defence.git}.

The evolution and advances made in the field of Cloud engineering influence the constant changes in software application development cycle and practices. Software architecture has evolved along with other domains and capabilities of software engineering. As migrating applications into the cloud is universally adopted by the software industry, microservices have proven to be the most suitable and widely accepted architecture pattern for applications deployed on distributed cloud. Their efficacy is enabled by both technical benefits like reliability, fault isolation, scalability and productivity benefits like ease of asset maintenance and clear ownership boundaries which in turn lead to fewer interdependencies and shorter development cycles thereby resulting in faster time to market. Though microservices have been established as an architecture pattern over the last decade, many organizations fail to optimize the architecture design to maximize efficiency. In some cases, the complexity of migrating an existing application into the microservices architecture becomes overwhelmingly complex and expensive. Additionally, automation and tool support for this problem are still at an early stage as there isn't a single well-acknowledged pattern or tool which could support the decomposition. This paper discusses a few impactful previous research and survey efforts to identify the lack of infrastructure cost optimization as a parameter in any of the approaches present. This paper proposes an Infrastructure-optimised predictive algorithm for partitioning monolithic software into microservices. It also summarizes the scope for future research opportunities within the area of microservices architecture and distributed cloud networks., Comment: 9 pages

The Indian Institute of Astrophysics (IIA) is developing a Multi-Conjugate Adaptive Optics (MCAO) system for the Kodaikanal Tower Telescope (KTT). In this context, we have measured the daytime turbulence strength profile at the Kodaikanal Observatory. The first method based on wavefront sensor (WFS) images, called S-DIMM+ (Solar-Differential Image Motion Monitor+), was used to estimate the higher altitude turbulence up to a height of 5 - 6 km. The second method used balloon-borne temperature sensors to measure the near-Earth turbulence up to 350 m. We also carried out simulations to validate the performance of our system. We report the first-ever daytime turbulence strength profile measurements at the observatory. We have identified the presence of a strong turbulence layer about 3 km above the observatory. The measured near-Earth turbulence matches the trend that is expected from the model for daytime component of turbulence and gives an integrated $r_0$ of about 4 cm at 500 nm. This is consistent with earlier seeing measurements. This shows that a low-cost setup with a small telescope and a simple array of temperature sensors can be used for estimating the turbulence strength profile at the site., Comment: Accepted for publication in JATIS

This work studies the product and ratio statistics of independent and non-identically distributed (i.n.i.d) $ \alpha-\kappa - \mu $ shadowed random variables. We derive the series expression for the probability density function (PDF), cumulative distribution function (CDF), and moment generating function (MGF) of the product and ratio of i.n.i.d $ \alpha - \kappa - \mu $ shadowed random variables. We then give the single integral representation for the derived PDF expressions. Further, as application examples, 1) outage probability has been derived for cascaded wireless systems, and 2) physical-layer security metrics like secrecy outage probability and strictly positive secrecy capacity are derived for the classic three-node model with $\alpha-\kappa-\mu$ shadowed fading. Next, we discuss an intelligent reflecting surface-assisted communication system over $\alpha-\kappa-\mu$ shadowed fading., Comment: arXiv admin note: text overlap with arXiv:2203.15760

Let $K$ be a finite unramified extension of $\mathbb{Q}_p$ with $p > 5$. We study the extremely non-generic irreducible components in the reduced part of the Emerton-Gee stack for $\mathrm{GL}_2$. We show precisely which irreducible components are smooth or normal, and which have Gorenstein or Cohen-Macaulay normalizations, as well as give a description of the singular loci. We use our results to update expectations about the conjectural categorical $p$-adic Langlands correspondence., Comment: 44 pages

Learning rate schedulers have shown great success in speeding up the convergence of learning algorithms in practice. However, their convergence to a minimum has not been proven theoretically. This difficulty mainly arises from the fact that, while traditional convergence analysis prescribes to monotonically decreasing (or constant) learning rates, schedulers opt for rates that often increase and decrease through the training epochs. In this work, we aim to bridge the gap by proposing a probabilistic learning rate scheduler (PLRS), that does not conform to the monotonically decreasing condition, with provable convergence guarantees. In addition to providing detailed convergence proofs, we also show experimental results where the proposed PLRS performs competitively as other state-of-the-art learning rate schedulers across a variety of datasets and architectures.

In this work, we propose a method to improve the energy efficiency and fairness of simultaneously transmitting and reflecting reconfigurable intelligent surfaces (STAR-RIS) for mobile users, ensuring reduced power consumption while maintaining reliable communication. To achieve this, we introduce a new parameter known as the subsurface assignment variable, which determines the number of STAR-RIS elements allocated to each user. We then formulate a novel optimization problem by concurrently optimizing the phase shifts of the STAR-RIS and subsurface assignment variable. We leverage the deep reinforcement learning (DRL) technique to address this optimization problem. The DRL model predicts the phase shifts of the STAR-RIS and efficiently allocates elements of STAR-RIS to the users. Additionally, we incorporate a penalty term in the DRL model to facilitate intelligent deactivation of STAR-RIS elements when not in use to enhance energy efficiency. Through extensive experiments, we show that the proposed method can achieve fairly high and nearly equal data rates for all users in both the transmission and reflection spaces in an energy-efficient manner.

Data mining has various real-time applications in fields such as finance telecommunications, biology, and government. Classification is a primary task in data mining. With the rise of cloud computing, users can outsource and access their data from anywhere, offloading data and it is processing to the cloud. However, in public cloud environments while data is often encrypted, the cloud service provider typically controls the encryption keys, meaning they can potentially access the data at any time. This situation makes traditional privacy-preserving classification systems inadequate. The recommended protocol ensures data privacy, protects user queries, and conceals access patterns. Given that encrypted data on the cloud cannot be directly mined, we focus on a secure k nearest neighbor classification algorithm for encrypted, outsourced data. This approach maintains the privacy of user queries and data access patterns while allowing effective data mining operations to be conducted securely in the cloud. With cloud computing, particularly in public cloud environments, the encryption of data necessitates advanced methods like secure k nearest neighbor algorithms to ensure privacy and functionality in data mining. This innovation protects sensitive information and user privacy, addressing the challenges posed by traditional systems where cloud providers control encryption keys.

We introduce DataComp for Language Models (DCLM), a testbed for controlled dataset experiments with the goal of improving language models. As part of DCLM, we provide a standardized corpus of 240T tokens extracted from Common Crawl, effective pretraining recipes based on the OpenLM framework, and a broad suite of 53 downstream evaluations. Participants in the DCLM benchmark can experiment with data curation strategies such as deduplication, filtering, and data mixing at model scales ranging from 412M to 7B parameters. As a baseline for DCLM, we conduct extensive experiments and find that model-based filtering is key to assembling a high-quality training set. The resulting dataset, DCLM-Baseline enables training a 7B parameter language model from scratch to 64% 5-shot accuracy on MMLU with 2.6T training tokens. Compared to MAP-Neo, the previous state-of-the-art in open-data language models, DCLM-Baseline represents a 6.6 percentage point improvement on MMLU while being trained with 40% less compute. Our baseline model is also comparable to Mistral-7B-v0.3 and Llama 3 8B on MMLU (63% & 66%), and performs similarly on an average of 53 natural language understanding tasks while being trained with 6.6x less compute than Llama 3 8B. Our results highlight the importance of dataset design for training language models and offer a starting point for further research on data curation., Comment: Project page: https://www.datacomp.ai/dclm/

Collisionless absorption of laser energy by an electron via laser-cluster interaction in an ambient magnetic field ($B_0$) has recently renewed interest. %due to high levels of absorption. Previously, using a rigid sphere model (RSM) and an extensive particle-in-cell (PIC) simulation with linearly polarized (LP) laser light, we have shown that an auxiliary field $B_0$ in a transverse direction to the laser polarization significantly enhances the laser absorption [Scientific Reports {\bf 12}, 11256 (2022)]. In this LP case, the average energy ($\mathcal{E}_A$) of an electron rises near $30-70$ times of its ponderomotive energy ($U_p$). The two-stage laser absorption by cluster electrons has been attributed via anharmonic resonance (AHR) followed by electron-cyclotron resonance (ECR) satisfying the improved phase-matching and frequency-matching conditions simultaneously. %Again, we report the narrow cone-like propagation of these %energetic electron bunches as a weakly relativistic electron beam with an %angular spread $\Delta\theta<5^{\circ}$. %Also, we highlight the effect of bigger cluster sizes on the energy and %angular spread of these electrons. In the present work, we study the effect of circularly polarized (CP) laser fields on the cluster-electron dynamics considering left/right circular polarizations with an ambient $B_0$. %, energy distribution, and..

We try to find the possibility of a Bianchi V universe in the modified gravitational field theory of $f(R,T)$. We have considered a Lagrangian model in the connection between the trace of the energy-momentum tensor $T$ and the Ricci scalar $R$. In order to solve the field equations a power law for the scaling factor was also considered. To make a comparison of the model parameters with the observational data, we put constraints on the model under the datasets of the Hubble parameter, Baryon Acoustic Oscillations, Pantheon, joint datasets of Hubble parameter + Pantheon, and collective datasets of the Hubble parameter + Baryon Acoustic Oscillations + Pantheon. The outcomes for the Hubble parameter in the present epoch are reasonably acceptable, especially since our estimation of this $H_0$ is remarkably consistent with various recent Planck Collaboration studies that utilize the $\Lambda$-CDM model.

This paper investigates the performance of multiple reconfigurable intelligent surfaces (multi-RIS) communication systems where the RIS link with the highest signal-to-noise-ratio (SNR) is selected at the destination. In practice, all the RISs will not have the same number of reflecting elements. Hence, selecting the RIS link with the highest SNR will involve characterizing the distribution of the maximum of independent, non-identically distributed (i.n.i.d.) SNR random variables (RVs). Using extreme value theory (EVT), we derive the asymptotic distribution of the normalized maximum of i.n.i.d. non-central chi-square (NCCS) distributed SNR RVs with one degree of freedom (d.o.f) and then extend the results for k-th order statistics. Using these asymptotic results, the outage capacity and average throughput expressions are derived for the multi-RIS system. The results for independent and identically distributed (i.i.d.) SNR RVs are then derived as a special case of i.n.i.d. RVs. All the derivations are validated through extensive Monte Carlo simulations, and their utility is discussed.

Recent Large Language Models (LLMs) have demonstrated remarkable capabilities in generating text that closely resembles human writing across wide range of styles and genres. However, such capabilities are prone to potential abuse, such as fake news generation, spam email creation, and misuse in academic assignments. Hence, it is essential to build automated approaches capable of distinguishing between artificially generated text and human-authored text. In this paper, we propose a simple yet efficient solution to this problem by ensembling predictions from multiple constituent LLMs. Compared to previous state-of-the-art approaches, which are perplexity-based or uses ensembles with a number of LLMs, our condensed ensembling approach uses only two constituent LLMs to achieve comparable performance. Experiments conducted on four benchmark datasets for generative text classification show performance improvements in the range of 0.5 to 100\% compared to previous state-of-the-art approaches. We also study the influence that the training data from individual LLMs have on model performance. We found that substituting commercially-restrictive Generative Pre-trained Transformer (GPT) data with data generated from other open language models such as Falcon, Large Language Model Meta AI (LLaMA2), and Mosaic Pretrained Transformers (MPT) is a feasible alternative when developing generative text detectors. Furthermore, to demonstrate zero-shot generalization, we experimented with an English essays dataset, and results suggest that our ensembling approach can handle new data effectively.

This paper examines a power law solution under $f(R,T)$ gravity for an isotropic and homogeneous universe by considering its functional form as $f(R,T) = R + \xi RT$, where $\xi$ is a positive constant. In $f(R,T)$ gravity, we have built the field equation for homogeneous and isotropic spacetime. The developed model's solution is $a = \alpha t^{\beta}$. We have used the redshift in the range $0 \leq z \leq 1.965$ and obtained the model parameters $\alpha$, $\beta$, $H_0$ by using the Markov Chain Monte Carlo (MCMC) method. The constrained values of the model parameter are as follows: $H_0 = 67.098^{+2.148}_{-1.792}$ km s$^{-1}$ Mpc$^{-1}$, $H_0 = 67.588^{+2.229}_{-2.170}$ km s$^{-1}$ Mpc$^{-1}$, $H_0 = 66.270^{+2.215}_{-2.181}$ km s$^{-1}$ Mpc$^{-1}$, $H_0 = 65.960^{+2.380}_{-1.834}$ km s$^{-1}$ Mpc$^{-1}$, $H_0 = 66.274^{+2.015}_{-1.864}$ km s$^{-1}$ Mpc$^{-1}$ which have been achieved by bounding the model with the Hubble parameter ($H(z)$) dataset, Baryon Acoustic Oscillations (BAO) dataset, Pantheon dataset, joint $H(z)$ + Pantheon dataset and collective $H(z)$ + BAO + Pantheon dataset, respectively. These computed $H_o$ observational values agree well with the outcomes from the Plank collaboration group. Through an analysis of the energy conditions' behaviour on our obtained solution, the model has been examined and analysed. Using the Om diagnostic as the state finder diagnostic tool and the jerk parameter, we have also investigated the model's validity. Our results show that, within a certain range of restrictions, the proposed model agrees with the observed signatures.

Adaptive Optics (AO) systems have become integral for ground-based astronomy. Based on the scientific case, there are various flavours of AO systems. Measuring the turbulence strength profile ($C_N^2(h)$) and other site characteristics is essential before selecting a site or implementing certain types of AO systems. We used an iterative deconvolution procedure on long-exposure H-$\alpha$ images of the Sun to determine the isoplanatic patch size during the daytime. Then, we determined the relationship between turbulence along different directions and also obtained an analytical estimate of the $C_N^2(h)$ profile., Comment: Proceeding for AO4ELT7

Our objective is to derive the range and velocity of multiple targets from the delay-Doppler domain for radar sensing using orthogonal time frequency space (OTFS) signaling. Noise contamination affects the performance of OTFS signals in real-world environments, making radar sensing challenging. This work introduces a two-stage approach to tackle this issue. In the first stage, we use a generative adversarial network to denoise the corrupted OTFS samples, significantly improving the data quality. Following this, the denoised signals are passed to a convolutional neural network model to predict the values of the velocities and ranges of multiple targets. The proposed two-stage approach can predict the range and velocity of multiple targets, even in very low signal-to-noise ratio scenarios, with high accuracy and outperforms existing methods.

In the domain of Natural Language Processing (NLP), Named Entity Recognition (NER) stands out as a pivotal mechanism for extracting structured insights from unstructured text. This manuscript offers an exhaustive exploration into the evolving landscape of NER methodologies, blending foundational principles with contemporary AI advancements. Beginning with the rudimentary concepts of NER, the study spans a spectrum of techniques from traditional rule-based strategies to the contemporary marvels of transformer architectures, particularly highlighting integrations such as BERT with LSTM and CNN. The narrative accentuates domain-specific NER models, tailored for intricate areas like finance, legal, and healthcare, emphasizing their specialized adaptability. Additionally, the research delves into cutting-edge paradigms including reinforcement learning, innovative constructs like E-NER, and the interplay of Optical Character Recognition (OCR) in augmenting NER capabilities. Grounding its insights in practical realms, the paper sheds light on the indispensable role of NER in sectors like finance and biomedicine, addressing the unique challenges they present. The conclusion outlines open challenges and avenues, marking this work as a comprehensive guide for those delving into NER research and applications.

Let K/Q_p be unramified. Inside the Emerton-Gee stack X_2, one can consider the locus of two-dimensional mod p representations of the absolute Galois group of K having a crystalline lift with specified Hodge-Tate weights. We study the case where the Hodge-Tate weights are irregular, which is an analogue for Galois representations of the partial weight one condition for Hilbert modular forms. We prove that if the gap between each pair of weights is bounded by p (the irregular analogue of a Serre weight), then this locus is irreducible. We also establish various inclusion relations between these loci., Comment: 36 pages

We introduce the UT Campus Object Dataset (CODa), a mobile robot egocentric perception dataset collected on the University of Texas Austin Campus. Our dataset contains 8.5 hours of multimodal sensor data: synchronized 3D point clouds and stereo RGB video from a 128-channel 3D LiDAR and two 1.25MP RGB cameras at 10 fps; RGB-D videos from an additional 0.5MP sensor at 7 fps, and a 9-DOF IMU sensor at 40 Hz. We provide 58 minutes of ground-truth annotations containing 1.3 million 3D bounding boxes with instance IDs for 53 semantic classes, 5000 frames of 3D semantic annotations for urban terrain, and pseudo-ground truth localization. We repeatedly traverse identical geographic locations for a wide range of indoor and outdoor areas, weather conditions, and times of the day. Using CODa, we empirically demonstrate that: 1) 3D object detection performance in urban settings is significantly higher when trained using CODa compared to existing datasets even when employing state-of-the-art domain adaptation approaches, 2) sensor-specific fine-tuning improves 3D object detection accuracy and 3) pretraining on CODa improves cross-dataset 3D object detection performance in urban settings compared to pretraining on AV datasets. Using our dataset and annotations, we release benchmarks for 3D object detection and 3D semantic segmentation using established metrics. In the future, the CODa benchmark will include additional tasks like unsupervised object discovery and re-identification. We publicly release CODa on the Texas Data Repository, pre-trained models, dataset development package, and interactive dataset viewer on our website at https://amrl.cs.utexas.edu/coda. We expect CODa to be a valuable dataset for research in egocentric 3D perception and planning for autonomous navigation in urban environments., Comment: 19 pages, 18 figures, 12 tables

Quantum Support Vector Machines (QSVM) play a vital role in using quantum resources for supervised machine learning tasks, such as classification. However, current methods are strongly limited in terms of scalability on Noisy Intermediate Scale Quantum (NISQ) devices. In this work, we propose a novel approach called the Variational Quantum Linear Solver (VQLS) enhanced QSVM. This is built upon our idea of utilizing the variational quantum linear solver to solve system of linear equations of a least squares-SVM on a NISQ device. The implementation of our approach is evaluated by an extensive series of numerical experiments with the Iris dataset, which consists of three distinct iris plant species. Based on this, we explore the practicality and effectiveness of our algorithm by constructing a classifier capable of classification in a feature space ranging from one to seven dimensions. Furthermore, by strategically exploiting both classical and quantum computing for various subroutines of our algorithm, we effectively mitigate practical challenges associated with the implementation. These include significant improvement in the trainability of the variational ansatz and notable reductions in run-time for cost calculations. Based on the numerical experiments, our approach exhibits the capability of identifying a separating hyperplane in an 8-dimensional feature space. Moreover, it consistently demonstrated strong performance across various instances with the same dataset.

The second largest eigenvalue of a graph is an important algebraic parameter which is related with the expansion, connectivity and randomness properties of a graph. Expanders are highly connected sparse graphs. In coding theory, Expander codes are Error Correcting codes made up of bipartite expander graphs. In this paper, first we prove the interlacing of the eigenvalues of the adjacency matrix of the bipartite graph with the eigenvalues of the bipartite quotient matrices of the corresponding graph matrices. Then we obtain bounds for the second largest and second smallest eigenvalues. Since the graph is bipartite, the results for Laplacian will also hold for Signless Laplacian matrix. We then introduce a new method called vertex-split of a bipartite graph to construct asymptotically good expander codes with expansion factor $\frac{D}{2}<\alpha < D$ and $\epsilon<\frac{1}{2}$ and prove a condition for the vertex-split of a bipartite graph to be $k-$connected with respect to $\lambda_{2}.$ Further, we prove that the vertex-split of $G$ is a bipartite expander. Finally, we construct an asymptotically good expander code whose factor graph is a graph obtained by the vertex-split of a bipartite graph., Comment: 17 pages, 2 figures

We introduce OpenFlamingo, a family of autoregressive vision-language models ranging from 3B to 9B parameters. OpenFlamingo is an ongoing effort to produce an open-source replication of DeepMind's Flamingo models. On seven vision-language datasets, OpenFlamingo models average between 80 - 89% of corresponding Flamingo performance. This technical report describes our models, training data, hyperparameters, and evaluation suite. We share our models and code at https://github.com/mlfoundations/open_flamingo.

The American Statistical Association Biopharmaceutical Section (ASA BIOP) working group on real-world evidence (RWE) has been making continuous, extended effort towards a goal of supporting and advancing regulatory science with respect to non-interventional, clinical studies intended to use real-world data for evidence generation for the purpose of medical product development and evaluation (i.e., RWE studies). In 2023, the working group published a manuscript delineating challenges and opportunities in constructing estimands for RWE studies following a framework in ICH E9(R1) guidance on estimand and sensitivity analysis. As a follow-up task, we describe the other issue in RWE studies, sensitivity analysis. Focusing on the issue of unmeasured confounding, we review availability and applicability of sensitivity analysis methods for different types unmeasured confounding. We discuss consideration on the choice and use of sensitivity analysis for RWE studies. Updated version of this article will present how findings from sensitivity analysis could support regulatory decision-making using a real example., Comment: 17 pages, 2 figures

Dense pixel-specific representation learning at scale has been bottlenecked due to the unavailability of large-scale multi-view datasets. Current methods for building effective pretraining datasets heavily rely on annotated 3D meshes, point clouds, and camera parameters from simulated environments, preventing them from building datasets from real-world data sources where such metadata is lacking. We propose a pretraining dataset-curation approach that does not require any additional annotations. Our method allows us to generate multi-view datasets from both real-world videos and simulated environments at scale. Specifically, we experiment with two scales: MIMIC-1M with 1.3M and MIMIC-3M with 3.1M multi-view image pairs. We train multiple models with different masked image modeling objectives to showcase the following findings: Representations trained on our automatically generated MIMIC-3M outperform those learned from expensive crowdsourced datasets (ImageNet-1K) and those learned from synthetic environments (MULTIVIEW-HABITAT) on two dense geometric tasks: depth estimation on NYUv2 (1.7%), and surface normals estimation on Taskonomy (2.05%). For dense tasks which also require object understanding, we outperform MULTIVIEW-HABITAT, on semantic segmentation on ADE20K (3.89%), pose estimation on MSCOCO (9.4%), and reduce the gap with models pre-trained on the object-centric expensive ImageNet-1K. We outperform even when the representations are frozen, and when downstream training data is limited to few-shot. Larger dataset (MIMIC-3M) significantly improves performance, which is promising since our curation method can arbitrarily scale to produce even larger datasets. MIMIC code, dataset, and pretrained models are open-sourced at https://github.com/RAIVNLab/MIMIC.

Multimodal datasets are a critical component in recent breakthroughs such as Stable Diffusion and GPT-4, yet their design does not receive the same research attention as model architectures or training algorithms. To address this shortcoming in the ML ecosystem, we introduce DataComp, a testbed for dataset experiments centered around a new candidate pool of 12.8 billion image-text pairs from Common Crawl. Participants in our benchmark design new filtering techniques or curate new data sources and then evaluate their new dataset by running our standardized CLIP training code and testing the resulting model on 38 downstream test sets. Our benchmark consists of multiple compute scales spanning four orders of magnitude, which enables the study of scaling trends and makes the benchmark accessible to researchers with varying resources. Our baseline experiments show that the DataComp workflow leads to better training sets. In particular, our best baseline, DataComp-1B, enables training a CLIP ViT-L/14 from scratch to 79.2% zero-shot accuracy on ImageNet, outperforming OpenAI's CLIP ViT-L/14 by 3.7 percentage points while using the same training procedure and compute. We release DataComp and all accompanying code at www.datacomp.ai., Comment: NeurIPS 2023 Datasets and Benchmarks Track

Twenty-six years ago a small committee report, building on earlier studies, expounded a compelling and poetic vision for the future of astronomy, calling for an infrared-optimized space telescope with an aperture of at least $4m$. With the support of their governments in the US, Europe, and Canada, 20,000 people realized that vision as the $6.5m$ James Webb Space Telescope. A generation of astronomers will celebrate their accomplishments for the life of the mission, potentially as long as 20 years, and beyond. This report and the scientific discoveries that follow are extended thank-you notes to the 20,000 team members. The telescope is working perfectly, with much better image quality than expected. In this and accompanying papers, we give a brief history, describe the observatory, outline its objectives and current observing program, and discuss the inventions and people who made it possible. We cite detailed reports on the design and the measured performance on orbit., Comment: Accepted by PASP for the special issue on The James Webb Space Telescope Overview, 29 pages, 4 figures

This paper deals with speeding up the convergence of a class of two-step iterative methods for solving linear systems of equations. To implement the acceleration technique, the residual norm associated with computed approximations for each sub-iterate is minimized over a certain two-dimensional subspace. Convergence properties of the proposed method are studied in detail. The approach is further developed to solve (regularized) normal equations arising from the discretization of ill-posed problems. The results of numerical experiments are reported to illustrate the performance of exact and inexact variants of the method on several test problems from different application areas.

Recent studies [Sci Rep 12, 11256 (2022)] on laser interaction (wavelength 800~nm, intensity $>10^{16}\, \Wcmcm$) with deuterium nano-cluster in an ambient magnetic field ($B_0$) demonstrate that collisionless absorption of laser occurs in two stages via anharmonic resonance (AHR) and electron-cyclotron resonance (ECR) or relativistic ECR (RECR) processes. Auxiliary $B_0$ enhances coupling of laser to cluster-electrons via improved frequency-matching for ECR/RECR as well as phase-matching for prolonged duration of the 5-fs (fwhm) broadband pulse and the average absorbed energy per electron $\overline{\mathcal{E}}_A$ significantly jumps up $\approx 36-70$ times of its ponderomotive energy ($\Up$). In this paper, we report energy dispersion of these energetic electrons and their angular distribution in position and momentum space by performing hybrid-PIC simulations. By simulating bigger clusters (radius $R_0 \approx 3-4$~nm) at high intensities $\approx 10^{16} - 10^{18}\,\Wcmcm$, we find $\overline{\mathcal{E}}_A\approx 36-70\,\Up$ similar to a small cluster ($R_0\approx 2$~nm), but total energy absorption increases almost linearly with increasing cluster size due to more number of available energy carriers.

Conventionally, in a differentially private additive noise mechanism, independent and identically distributed (i.i.d.) noise samples are added to each coordinate of the response. In this work, we formally present the addition of noise which is independent, but not identically distributed (i.n.i.d.) across the coordinates to achieve tighter privacy-accuracy trade-off by exploiting coordinate-wise disparity. In particular, we study the i.n.i.d. Gaussian and Laplace mechanisms and obtain the conditions under which these mechanisms guarantee privacy. The optimal choice of parameters that ensure these conditions are derived theoretically. Theoretical analyses and numerical simulations demonstrate that the i.n.i.d. mechanisms achieve higher utility for the given privacy requirements compared to their i.i.d. counterparts. One of the interesting observations is that the Laplace mechanism outperforms Gaussian even in high dimensions, as opposed to the popular belief, if the irregularity in coordinate-wise sensitivities is exploited. We also demonstrate how the i.n.i.d. noise can improve the performance in the private empirical risk minimization through coordinate descent.

Modern large ground-based solar telescopes are invariably equipped with adaptive optics systems to enhance the high angular resolution imaging and spectroscopic capabilities in the presence of the Earth's atmospheric turbulence. The quality of the images obtained from these telescopes can not be quantified with the Strehl ratio or other metrics that are used for nighttime astronomical telescopes directly. In this paper, we propose to use the root mean square (rms) granulation contrast as a metric to quantify the image quality of ground-based solar telescopes. We obtain semi-logarithmic plots indicating the correspondence between the Strehl ratio and the rms granulation contrast for most practical values of the telescope diameters (D) and the atmospheric coherence diameters ($ r_0$), for various levels of adaptive optics compensation. We estimate the efficiency of a few working solar adaptive optics systems by comparing the results of our simulations with the Strehl ratio and rms granulation contrast published by these systems. Our results can be used in conjunction with a plausible 50 system efficiency to predict the lower bound on the rms granulation contrast expected from ground-based solar telescopes., Comment: 15 pages, 8 figures

We propose a deep reinforcement learning (DRL) approach for a full-duplex (FD) transmission that predicts the phase shifts of the reconfigurable intelligent surface (RIS), base station (BS) active beamformers, and the transmit powers to maximize the weighted sum rate of uplink and downlink users. Existing methods require channel state information (CSI) and residual self-interference (SI) knowledge to calculate exact active beamformers or the DRL rewards, which typically fail without CSI or residual SI. Especially for time-varying channels, estimating and signaling CSI to the DRL agent is required at each time step and is costly. We propose a two-stage DRL framework with minimal signaling overhead to address this. The first stage uses the least squares method to initiate learning by partially canceling the residual SI. The second stage uses DRL to achieve performance comparable to existing CSI-based methods without requiring the CSI or the exact residual SI. Further, the proposed DRL framework for quantized RIS phase shifts reduces the signaling from BS to the RISs using $32$ times fewer bits than the continuous version. The quantized methods reduce action space, resulting in faster convergence and $7.1\%$ and $22.28\%$ better UL and DL rates, respectively than the continuous method.

The framework of differential privacy protects an individual's privacy while publishing query responses on congregated data. In this work, a new noise addition mechanism for differential privacy is introduced where the noise added is sampled from a hybrid density that resembles Laplace in the centre and Gaussian in the tail. With a sharper centre and light, sub-Gaussian tail, this density has the best characteristics of both distributions. We theoretically analyze the proposed mechanism, and we derive the necessary and sufficient condition in one dimension and a sufficient condition in high dimensions for the mechanism to guarantee (${\epsilon}$,${\delta}$)-differential privacy. Numerical simulations corroborate the efficacy of the proposed mechanism compared to other existing mechanisms in achieving a better trade-off between privacy and accuracy.

Quantum state tomography aims to estimate the state of a quantum mechanical system which is described by a trace one, Hermitian positive semidefinite complex matrix, given a set of measurements of the state. Existing works focus on estimating the density matrix that represents the state, using a compressive sensing approach, with only fewer measurements than that required for a tomographically complete set, with the assumption that the true state has a low rank. One very popular method to estimate the state is the use of the Singular Value Thresholding (SVT) algorithm. In this work, we present a machine learning approach to estimate the quantum state of n-qubit systems by unrolling the iterations of SVT which we call Learned Quantum State Tomography (LQST). As merely unrolling SVT may not ensure that the output of the network meets the constraints required for a quantum state, we design and train a custom neural network whose architecture is inspired from the iterations of SVT with additional layers to meet the required constraints. We show that our proposed LQST with very few layers reconstructs the density matrix with much better fidelity than the SVT algorithm which takes many hundreds of iterations to converge. We also demonstrate the reconstruction of the quantum Bell state from an informationally incomplete set of noisy measurements., Comment: A preliminary version of this paper, [1] is uploaded to arXiv which is not tailored to quantum state estimation. The paper [1] deals with the estimation of a general low-rank real matrix whereas the current paper is tailored for estimating the quantum state of n-qubit systems and the current paper is applied to estimating the Bell state from its noisy measurements. [1] arXiv:2105.06934

We design the phase shifts of an intelligent reflecting surface (IRS)-assisted single-input-multiple-output communication system to minimize the outage probability (OP) and to maximize the ergodic rate. Our phase shifts design uses only statistical channel state information since these depend only on the large-scale fading coefficients; the obtained phase shift design remains valid for a longer time frame. We further assume that one has access to only quantized phase values. The closed-form expressions for OP and ergodic rate are derived for the considered system. Next, two optimization problems are formulated to choose the phase shifts of IRS such that (i) OP is minimized and (ii) the ergodic rate is maximized. We used the multi-valued particle swarm optimization (MPSO) and particle swarm optimization (PSO) algorithms to solve the optimization problems. Numerical simulations are performed to study the impact of various parameters on the OP and ergodic rate. We also discuss signaling overhead between BS and IRS controller. It is shown that the overhead can be reduced up to $99.69 \%$ by using statistical CSI for phase shift design and $5$ bits to represent the phase shifts without significantly compromising on the performance.

We introduce the Text Classification Attack Benchmark (TCAB), a dataset for analyzing, understanding, detecting, and labeling adversarial attacks against text classifiers. TCAB includes 1.5 million attack instances, generated by twelve adversarial attacks targeting three classifiers trained on six source datasets for sentiment analysis and abuse detection in English. Unlike standard text classification, text attacks must be understood in the context of the target classifier that is being attacked, and thus features of the target classifier are important as well. TCAB includes all attack instances that are successful in flipping the predicted label; a subset of the attacks are also labeled by human annotators to determine how frequently the primary semantics are preserved. The process of generating attacks is automated, so that TCAB can easily be extended to incorporate new text attacks and better classifiers as they are developed. In addition to the primary tasks of detecting and labeling attacks, TCAB can also be used for attack localization, attack target labeling, and attack characterization. TCAB code and dataset are available at https://react-nlp.github.io/tcab/., Comment: 32 pages, 7 figures, and 14 tables

Let $p$ be a fixed odd prime, and let $K$ be a finite extension of $\mathbb{Q}_p$ with ring of integers $\mathcal{O}_K$. The Emerton-Gee stack for $\mathrm{GL}_2$ is a stack of $(\varphi, \Gamma)$-modules. The stack, denoted $\mathcal{X}_2$, can be interpreted as a moduli stack of representations of the absolute Galois group of $K$ with $p$-adic coefficients. The reduced part of the Emerton-Gee stack, denoted $\mathcal{X}_{2, \text{red}}$, is an algebraic stack defined over a finite field of characteristic $p$ and can be viewed as a moduli stack of Galois representations with mod $p$ coefficients. The irreducible components of $\mathcal{X}_{2, \text{red}}$ are labelled in a natural way by Serre weights, which are the irreducible mod $p$ representations of $\mathrm{GL}_2(\mathcal{O}_K)$. Each irreducible component of $\mathcal{X}_{2, \text{red}}$ has dimension $[K:\mathbb{Q}_p]$. Motivated by the conjectural categorical $p$-adic Langlands programme, we find representation-theoretic criteria for codimension one intersections of the irreducible components of $\mathcal{X}_{2, \text{red}}$. The methods involve two separate computations and a final comparison between the two. The first of these computations determines extension groups of Serre weights and the second determines all the pairs of irreducible components that intersect in codimension one. We show that a non-trivial extension of a pair of non-isomorphic Serre weights implies a codimension one intersection of the corresponding irreducible components. The converse of this statement is also true when the Serre weights are chosen to be sufficiently generic. Furthermore, we show that the number of top-dimensional components in a codimension one intersection is related to the nature of the extension group of corresponding Serre weights., Comment: Did a significant rewrite of the article to improve readability, but without changes in actual results. Omitted calculations with p=2 since they were incomplete anyway and confused the presentation

Let $K$ be a finite unramified extension of $\mathbb{Q}_p$, where $p>2$. [CEGS22b] and [EG23] construct a moduli stack of two dimensional mod $p$ representations of the absolute Galois group of $K$. We show that most irreducible components of this stack (including several non-generic components) are isomorphic to quotients of smooth affine schemes. We also use this quotient presentation to compute global sections on these components., Comment: Final version. To appear in Transactions of the AMS

Reconfigurable intelligent surfaces (RIS) can be crucial in next-generation communication systems. However, designing the {RIS} phases according to the instantaneous channel state information (CSI) can be challenging in practice due to the short coherent time of the channel. In this regard, we propose a novel algorithm based on the channel statistics of massive multiple input multiple output systems rather than the instantaneous {CSI}. The beamforming at the base station (BS), power allocation of the users, and phase shifts at the RIS elements are optimized to maximize the minimum signal-to-interference and noise ratio (SINR), guaranteeing fair operation among various users. In particular, we design the RIS phases by leveraging the asymptotic deterministic equivalent of the minimum {SINR} that depends only on the channel statistics. This significantly reduces the computational complexity and the amount of controlling data between the {BS} and {RIS} for updating the phases. This setup is also useful for electromagnetic fields (EMF)-aware systems with constraints on the maximum user's exposure to EMF. The numerical results show that the proposed algorithms achieve more than $100 \%$ gain in terms of minimum SINR, compared to a system with random RIS phase shifts, when $40$ RIS elements, $20$ antennas at the BS and $10$ users, are considered.

Cohort Intelligence or CI is one of its kind of novel optimization algorithm. Since its inception, in a very short span it is applied successfully in various domains and its results are observed to be effectual in contrast to algorithm of its kind. Till date, there is no such type of bibliometric analysis carried out on CI and its related applications. So, this research paper in a way will be an ice breaker for those who want to take up CI to a new level. In this research papers, CI publications available in Scopus are analyzed through graphs, networked diagrams about authors, source titles, keywords over the years, journals over the time. In a way this bibliometric paper showcase CI, its applications and detail outs systematic review in terms its bibliometric details.

Attribute values of the products are an essential component in any e-commerce platform. Attribute Value Extraction (AVE) deals with extracting the attributes of a product and their values from its title or description. In this paper, we propose to tackle the AVE task using generative frameworks. We present two types of generative paradigms, namely, word sequence-based and positional sequence-based, by formulating the AVE task as a generation problem. We conduct experiments on two datasets where the generative approaches achieve the new state-of-the-art results. This shows that we can use the proposed framework for AVE tasks without additional tagging or task-specific model design., Comment: 6 pages

In this work, we characterize the outage probability (OP) of an intelligent reflecting surface (IRS) assisted multi-user multiple-input-single-output (MU-MISO) communication system. Using a two-step approximation method, we approximate the signal-to-interference-plus-noise ratio (SINR) for any downlink user by a Log-Normal random variable. The impact of various system parameters is studied using the closed-form expression of OP. It is concluded that the position of IRS has a critical role, but an appropriate increase in the number of IRS elements would help to compensate for the loss in performance if the position of IRS is suboptimal.

Smart radio environments aided by reconfigurable intelligent reflecting surfaces (RIS) have attracted much research attention recently. We propose a joint optimization strategy for beamforming, RIS phases, and power allocation to maximize the minimum SINR of an uplink RIS-aided communication system. The users are subject to constraints on their transmit power. We derive a closed-form expression for the beam forming vectors and a geometric programming-based solution for power allocation. We also propose two solutions for optimizing the phase shifts at the RIS, one based on the matrix lifting method and one using an approximation for the minimum function. We also propose a heuristic algorithm for optimizing quantized phase shift values. The proposed algorithms are of practical interest for systems with constraints on the maximum allowable electromagnetic field exposure. For instance, considering $24$-element RIS, $12$-antenna BS, and $6$ users, numerical results show that the proposed algorithm achieves close to $300 \%$ gain in terms of minimum SINR compared to a scheme with random RIS phases.

Performing low-rank matrix completion with sensitive user data calls for privacy-preserving approaches. In this work, we propose a novel noise addition mechanism for preserving differential privacy where the noise distribution is inspired by Huber loss, a well-known loss function in robust statistics. The proposed Huber mechanism is evaluated against existing differential privacy mechanisms while solving the matrix completion problem using the Alternating Least Squares approach. We also propose using the Iteratively Re-Weighted Least Squares algorithm to complete low-rank matrices and study the performance of different noise mechanisms in both synthetic and real datasets. We prove that the proposed mechanism achieves {\epsilon}-differential privacy similar to the Laplace mechanism. Furthermore, empirical results indicate that the Huber mechanism outperforms Laplacian and Gaussian in some cases and is comparable, otherwise., Comment: 13 pages