Search

INTRODUCTION: Diffusion-weighted magnetic resonance imaging (dMRI) is sensitive to the microstructural properties of brain tissues and shows great promise in detecting the effects of degenerative diseases. However, many approaches analyze single measures averaged over regions of interest without considering the underlying fiber geometry. METHODS: We propose a novel macrostructure-informed normative tractometry (MINT) framework to investigate how white matter (WM) microstructure and macrostructure are jointly altered in mild cognitive impairment (MCI) and dementia. We compared MINT-derived metrics with univariate diffusion tensor imaging (DTI) metrics to examine how fiber geometry may impact the interpretation of microstructure. RESULTS: In two multisite cohorts from North America and India, we find consistent patterns of microstructural and macrostructural anomalies implicated in MCI and dementia; we also rank diffusion metrics sensitivity to dementia. DISCUSSION: We show that MINT, by jointly modeling tract shape and microstructure, has the potential to disentangle and better interpret the effects of degenerative disease on the brains neural pathways. HIGHLIGHTS: Changes in diffusion tensor imaging metrics may be due to macroscopic changes. Normative models encode normal variability of diffusion metrics in healthy controls. Variational autoencoder applied on tractography can learn patterns of fiber geometry. WM microstructure and macrostructure are modeled with multivariate methods. Transfer learning uses pretraining and fine-tuning for increased efficiency.

Spatio-temporal coherency is a major challenge in synthesizing high quality videos, particularly in synthesizing human videos that contain rich global and local deformations. To resolve this challenge, previous approaches have resorted to different features in the generation process aimed at representing appearance and motion. However, in the absence of strict mechanisms to guarantee such disentanglement, a separation of motion from appearance has remained challenging, resulting in spatial distortions and temporal jittering that break the spatio-temporal coherency. Motivated by this, we here propose LEO, a novel framework for human video synthesis, placing emphasis on spatio-temporal coherency. Our key idea is to represent motion as a sequence of flow maps in the generation process, which inherently isolate motion from appearance. We implement this idea via a flow-based image animator and a Latent Motion Diffusion Model (LMDM). The former bridges a space of motion codes with the space of flow maps, and synthesizes video frames in a warp-and-inpaint manner. LMDM learns to capture motion prior in the training data by synthesizing sequences of motion codes. Extensive quantitative and qualitative analysis suggests that LEO significantly improves coherent synthesis of human videos over previous methods on the datasets TaichiHD, FaceForensics and CelebV-HQ. In addition, the effective disentanglement of appearance and motion in LEO allows for two additional tasks, namely infinite-length human video synthesis, as well as content-preserving video editing. Project page: https://wyhsirius.github.io/LEO-project/. [ABSTRACT FROM AUTHOR]

Recent advancements in technology for planning support have led to increased interest in participatory planning support systems (PPSSs). However, existing PPSSs often struggle to facilitate higher levels of public participation due to limitations in their practical usefulness. Emerging large language models (LLMs) like ChatGPT, along with artificial intelligence (AI) technologies such as deep generative methods, offer new opportunities to enhance PPSS, though this potential has yet to be fully explored. This study aims to address these gaps by integrating LLMs, specifically ChatGPT, into a new web-based PPSS platform. The platform operates as a multi-agent system with five key components: users, process, agents, knowledge, and tools. Stakeholders engage with ChatGPT-enabled personalized agents that are supported by a project-specific knowledge base. These agents understand user preferences, concerns, and needs, and call upon task agents, also powered by ChatGPT, to execute tasks, including applying deep generative tools that enable stakeholders to create their own designs. A workflow agent coordinates the overall process, facilitating the sharing of information, data, opinions, and designs to promote communication and build consensus among stakeholders. The platform was tested in a hypothetical sustainable urban regeneration case, the New Seollo Project in Seoul. Compared to the actual Seoullo project, which faced substantial criticism, the simulated results suggest the platform's potential to significantly improve participation and generate better design solutions. This new PPSS platform enhances usefulness by improving stakeholder communication and empowering the public to contribute comprehensive, inclusive, and creative solutions for sustainable urban development. [ABSTRACT FROM AUTHOR]

This paper reviews scenario generation techniques for modeling uncertainty in wind and photovoltaic (PV) power generation, a critical component as renewable energy integration into power systems grows. Scenario generation enables the simulation of variable power outputs under different weather conditions, serving as essential inputs for robust, stochastic, and distributionally robust optimization in system planning and operation. We categorize scenario generation methods into explicit and implicit approaches. Explicit methods rely on probabilistic assumptions and parameter estimation, which enable the interpretable yet parameterized modeling of power variability. Implicit methods, powered by deep learning models, offer data-driven scenario generation without predefined distributions, capturing complex temporal and spatial patterns in the renewable output. The review also addresses combined wind and PV power scenario generation, highlighting its importance for accurately reflecting correlated fluctuations in multi-site, interconnected systems. Finally, we address the limitations of scenario generation for wind and PV power integration planning and suggest future research directions. [ABSTRACT FROM AUTHOR]

Background: The findings of the 2023 AAPM Grand Challenge on Deep Generative Modeling for Learning Medical Image Statistics are reported in this Special Report. Purpose: The goal of this challenge was to promote the development of deep generative models for medical imaging and to emphasize the need for their domain‐relevant assessments via the analysis of relevant image statistics. Methods: As part of this Grand Challenge, a common training dataset and an evaluation procedure was developed for benchmarking deep generative models for medical image synthesis. To create the training dataset, an established 3D virtual breast phantom was adapted. The resulting dataset comprised about 108 000 images of size 512 ×$\times$ 512. For the evaluation of submissions to the Challenge, an ensemble of 10 000 DGM‐generated images from each submission was employed. The evaluation procedure consisted of two stages. In the first stage, a preliminary check for memorization and image quality (via the Fréchet Inception Distance [FID]) was performed. Submissions that passed the first stage were then evaluated for the reproducibility of image statistics corresponding to several feature families including texture, morphology, image moments, fractal statistics, and skeleton statistics. A summary measure in this feature space was employed to rank the submissions. Additional analyses of submissions was performed to assess DGM performance specific to individual feature families, the four classes in the training data, and also to identify various artifacts. Results: Fifty‐eight submissions from 12 unique users were received for this Challenge. Out of these 12 submissions, 9 submissions passed the first stage of evaluation and were eligible for ranking. The top‐ranked submission employed a conditional latent diffusion model, whereas the joint runners‐up employed a generative adversarial network, followed by another network for image superresolution. In general, we observed that the overall ranking of the top 9 submissions according to our evaluation method (i) did not match the FID‐based ranking, and (ii) differed with respect to individual feature families. Another important finding from our additional analyses was that different DGMs demonstrated similar kinds of artifacts. Conclusions: This Grand Challenge highlighted the need for domain‐specific evaluation to further DGM design as well as deployment. It also demonstrated that the specification of a DGM may differ depending on its intended use. [ABSTRACT FROM AUTHOR]

Sampling-based tolerance analysis is a powerful tool for evaluating the quality of functional products, but requires realistic manufacturing distributions. However, the determination of resulting manufacturing distributions is usually associated with a high financial and time expenditure, especially for novel technologies such as Additive Manufacturing. Usually, sampling techniques are used to reproduce the original distribution of manufacturing variations based on statistical moments. In most cases, simplifying assumptions are made for this purpose, potentially leading to an inadequate representation of the correlation between machine and process parameters in the resulting distribution. In the worst case, this can lead to a falsification of the tolerance analysis results. Aiming to address this challenge, this paper presents an approach to imitate real-world manufacturing distributions using generative Machine Learning techniques based on Deep Learning with small real data sets. This enables a realistic reproduction of quasi-real manufacturing distributions and omits conventional sampling techniques. The general procedure and its applicability are shown via illustrative use cases from the tolerancing domain. [ABSTRACT FROM AUTHOR]

Surrogate modeling is of great practical significance for parametric differential equation systems. In contrast to classical numerical methods, using physics-informed deep learning-based methods to construct simulators for such systems is a promising direction due to its potential to handle high dimensionality, which requires minimizing a loss over a training set of random samples. However, the random samples introduce statistical errors, which may become the dominant errors for the approximation of low-regularity and high-dimensional problems. In this work, we present a deep adaptive sampling method for surrogate modeling of low-regularity parametric differential equations and illustrate the necessity of adaptive sampling for constructing surrogate models. In the parametric setting, the residual loss function can be regarded as an unnormalized probability density function (PDF) of the spatial and parametric variables. In contrast to the non-parametric setting, factorized joint density models can be employed to alleviate the difficulties induced by the parametric space. The PDF is approximated by a deep generative model, from which new samples are generated and added to the training set. Since the new samples match the residual-induced distribution, the refined training set can further reduce the statistical error in the current approximate solution through variance reduction. We demonstrate the effectiveness of the proposed method with a series of numerical experiments, including the physics-informed operator learning problem, the parametric optimal control problem with geometrical parametrization, and the parametric lid-driven 2D cavity flow problem with a continuous range of Reynolds numbers from 100 to 3200. [ABSTRACT FROM AUTHOR]

The challenge of automatically synthesizing high-quality vector fonts, particularly for writing systems (e.g., Chinese) consisting of huge amounts of complex glyphs, remains unsolved. Existing font synthesis techniques fall into two categories: 1) methods that directly generate vector glyphs, and 2) methods that initially synthesize glyph images and then vectorize them. However, the first category often fails to construct complete and correct shapes for complex glyphs, while the latter struggles to efficiently synthesize high-resolution (i.e., 1024 × 1024 or higher) glyph images while preserving local details. In this paper, we introduce HFH-Font, a few-shot font synthesis method capable of efficiently generating high-resolution glyph images that can be converted into high-quality vector glyphs. More specifically, our method employs a diffusion model-based generative framework with component-aware conditioning to learn different levels of style information adaptable to varying input reference sizes. We also design a distillation module based on Score Distillation Sampling for 1-step fast inference, and a style-guided super-resolution module to refine and upscale low-resolution synthesis results. Extensive experiments, including a user study with professional font designers, have been conducted to demonstrate that our method significantly outperforms existing font synthesis approaches. Experimental results show that our method produces high-fidelity, high-resolution raster images which can be vectorized into high-quality vector fonts. Using our method, for the first time, large-scale Chinese vector fonts of a quality comparable to those manually created by professional font designers can be automatically generated. [ABSTRACT FROM AUTHOR]

Albeit the recent successful demonstrations of nanophotonic device designs leveraging data‐driven supervised learning methods, several challenges still remain. One of the primary constraints of these methods is their computational cost because they require a massive amount of highly time‐consuming full‐wave electromagnetic simulations. In this study, semi‐supervised learning methods are implemented to avoid this issue. Photonic crystal waveguide devices, which offer interesting light‐matter interactions such as the slow‐light effect and the filtering effect, are employed to validate the methodologies. To utilize unlabeled data within this framework, a novel deep generative model, the denoising diffusion probabilistic model is introduced. Next, a pseudo‐labeling process is introduced to assign synthetic labels to the unlabeled data. The performance of a semi‐supervised learning method is compared with a supervised learning scheme, demonstrating that the performance of the neural network can be enhanced without additional numerical calculations. This method holds extensive potential to elevate the efficiency of designing and characterizing future nanophotonic devices. [ABSTRACT FROM AUTHOR]

Deep generative models (DGMs) have the potential to revolutionize diagnostic imaging. Generative adversarial networks (GANs) are one kind of DGM which are widely employed. The overarching problem with deploying any sort of DGM in mission-critical applications is a lack of adequate and/or automatic means of assessing the domain-specific quality of generated images. In this work, we demonstrate several objective and human-interpretable tests of images output by two popular DGMs. These tests serve two goals: (i) ruling out DGMs for downstream, domain-specific applications, and (ii) quantifying hallucinations in the expected spatial context in DGM-generated images. The designed datasets are made public and the proposed tests could also serve as benchmarks and aid the prototyping of emerging DGMs. Although these tests are demonstrated on GANs, they can be employed as a benchmark for evaluating any DGM. Specifically, we designed several stochastic context models (SCMs) of distinct image features that can be recovered after generation by a trained DGM. Together, these SCMs encode features as per-image constraints in prevalence, position, intensity, and/or texture. Several of these features are high-order, algorithmic pixel-arrangement rules which are not readily expressed in covariance matrices. We designed and validated statistical classifiers to detect specific effects of the known arrangement rules. We then tested the rates at which two different DGMs correctly reproduced the feature context under a variety of training scenarios, and degrees of feature-class similarity. We found that ensembles of generated images can appear largely accurate visually, and show high accuracy in ensemble measures, while not exhibiting the known spatial arrangements. The main conclusion is that SCMs can be engineered, and serve as benchmarks, to quantify numerous per image errors, i.e. , hallucinations, that may not be captured in ensemble statistics but plausibly can affect subsequent use of the DGM-generated images. • Assessment of images generated by DGMs for diagnostic use is an open challenge. • Domain-relevant spatial context for testing DGMs can be encoded in stochastic models. • Stochastic context models can be employed to quantify DGM hallucinations. • Tests of per-realization context yield crucial information beyond ensemble measures. • The proposed method is independent of DGM architecture & enables model benchmarking. [ABSTRACT FROM AUTHOR]

Deep Generative Models (DGMs), as a state-of-the-art technology in the field of artificial intelligence, find extensive applications across various domains. However, their security concerns have increasingly gained prominence, particularly with regard to invisible backdoor attacks. Currently, most backdoor attack methods rely on visible backdoor triggers that are easily detectable and defendable against. Although some studies have explored invisible backdoor attacks, they often require parameter modifications and additions to the model generator, resulting in practical inconveniences. In this study, we aim to overcome these limitations by proposing a novel method for invisible backdoor attacks. We employ an encoder–decoder network to 'poison' the data during the preparation stage without modifying the model itself. Through meticulous design, the trigger remains visually undetectable, substantially enhancing attacker stealthiness and success rates. Consequently, this attack method poses a serious threat to the security of DGMs while presenting new challenges for security mechanisms. Therefore, we urge researchers to intensify their investigations into DGM security issues and collaboratively promote the healthy development of DGM security. [ABSTRACT FROM AUTHOR]

Traditional urban design often overlooks the synchronisation of human and ecological connectivities, typically favouring corridors for ecological continuity. Our study challenges this convention by introducing a computational design approach, meta-connectivity, leveraging the deep generative models performing cross-domain translation to integrate human–wildlife landscape connectivity in urban morphology amidst the planetary urbanisation. Utilising chained Pix2Pix models, our research illustrates a novel meta-connectivity design reasoning framework, combining landscape connectivity modelling with conditional reasoning based on deep generative models. This framework enables the adjustment of both human and wildlife landscape connectivities based on their correlative patterns in one single design process, guiding the rematerialisation of urban landscapes without the need for explicit prior ecological or urban data. Our empirical study in East London demonstrated the framework's efficacy in suggesting wildlife connectivity adjustments based on human connectivity metrics. The results demonstrate the feasibility of creating an innovative urban form in which the land cover guided by the connectivity gradients replaces the corridors based on simple geometries. This research thus presents a methodology shift in urban design, proposing a symbiotic approach to integrating disparate yet interrelated landscape connectivities within urban contexts. [ABSTRACT FROM AUTHOR]

With their diverse biological activities, peptides are promising candidates for therapeutic applications, showing antimicrobial, antitumour and hormonal signalling capabilities. Despite their advantages, therapeutic peptides face challenges such as short half-life, limited oral bioavailability and susceptibility to plasma degradation. The rise of computational tools and artificial intelligence (AI) in peptide research has spurred the development of advanced methodologies and databases that are pivotal in the exploration of these complex macromolecules. This perspective delves into integrating AI in peptide development, encompassing classifier methods, predictive systems and the avant-garde design facilitated by deep-generative models like generative adversarial networks and variational autoencoders. There are still challenges, such as the need for processing optimization and careful validation of predictive models. This work outlines traditional strategies for machine learning model construction and training techniques and proposes a comprehensive AI-assisted peptide design and validation pipeline. The evolving landscape of peptide design using AI is emphasized, showcasing the practicality of these methods in expediting the development and discovery of novel peptides within the context of peptide-based drug discovery. [ABSTRACT FROM AUTHOR]

Despite the proliferation of deep learning techniques for accelerated MRI acquisition and enhanced image reconstruction, the construction of large and diverse MRI datasets continues to pose a barrier to effective clinical translation of these technologies. One major challenge is in collecting the MRI raw data (required for image reconstruction) from clinical scanning, as only magnitude images are typically saved and used for clinical assessment and diagnosis. The image phase and multi-channel RF coil information are not retained when magnitude-only images are saved in clinical imaging archives. Additionally, preprocessing used for data in clinical imaging can lead to biased results. While several groups have begun concerted efforts to collect large amounts of MRI raw data, current databases are limited in the diversity of anatomy, pathology, annotations, and acquisition types they contain. To address this, we present a method for synthesizing realistic MR data from magnitude-only data, allowing for the use of diverse data from clinical imaging archives in advanced MRI reconstruction development. Our method uses a conditional GAN-based framework to generate synthetic phase images from input magnitude images. We then applied ESPIRiT to derive RF coil sensitivity maps from fully sampled real data to generate multi-coil data. The synthetic data generation method was evaluated by comparing image reconstruction results from training Variational Networks either with real data or synthetic data. We demonstrate that the Variational Network trained on synthetic MRI data from our method, consisting of GAN-derived synthetic phase and multi-coil information, outperformed Variational Networks trained on data with synthetic phase generated using current state-of-the-art methods. Additionally, we demonstrate that the Variational Networks trained with synthetic k-space data from our method perform comparably to image reconstruction networks trained on undersampled real k-space data.

In this article, we develop CausalEGM, a deep learning framework for nonlinear dimension reduction and generative modeling of the dependency among covariate features affecting treatment and response. CausalEGM can be used for estimating causal effects in both binary and continuous treatment settings. By learning a bidirectional transformation between the high-dimensional covariate space and a low-dimensional latent space and then modeling the dependencies of different subsets of the latent variables on the treatment and response, CausalEGM can extract the latent covariate features that affect both treatment and response. By conditioning on these features, one can mitigate the confounding effect of the high dimensional covariate on the estimation of the causal relation between treatment and response. In a series of experiments, the proposed method is shown to achieve superior performance over existing methods in both binary and continuous treatment settings. The improvement is substantial when the sample size is large and the covariate is of high dimension. Finally, we established excess risk bounds and consistency results for our method, and discuss how our approach is related to and improves upon other dimension reduction approaches in causal inference. [ABSTRACT FROM AUTHOR]

The exchange of information through social networking sites has become a major risk due to the possibility of obtaining millions of subscribers’ data at any time without the right. Multimedia security is a multifaceted field that involves various techniques and technologies to protect digital media in different contexts. As the technology evolves, so do the challenges and solutions related to multimedia security. Steganography plays a dominant role in covert communication over these social networking. In most modern adaptive steganography, the balancing between imperceptibility, payload, and security is a critical difficulty for image steganography. To this end, in this paper, we propose an improved image steganography method called IS-DGM based on a deep generative model (DGM) combined with hyper logistic map (HLM) encryption algorithm. IS-DGM consists of two strategies, steganography and recovery. In the first strategy, we have pre-processing and embedding networks. Before running the pre-processing network, the secret image is encoded using the HLM algorithm. During this phase, the encoded and the carrier images are utilized as inputs of the embedding network to boost concealment efficiency. In the second strategy, we have extraction and steganalysis networks. During this phase, the secret is extracted from the host image with the good visual quality as possible. Experimental outcomes indicate that the proposed method performs effectively in terms of perceptual quality and embedding capacity on five data sets, namely, ImageNet, CoCo2017, LFW, VoC2007, and VoC2012. In addition, it outperforms recent deep learning GAN hiding algorithms with respect to capacity, visual quality, and security. Thus, the proposed IS-DGM effectively balances good imperceptibility and increased capacity. Further, it maintains safety against histogram analysis, such as PVD analysis. Besides, the IS-DGM method increases resistance to the ROC curve analysis, including steganalysis algorithms, such as SRM, MaxSRM, Stegexpose, and Ye-Net. [ABSTRACT FROM AUTHOR]

Integrating deep generative models into urban form generation is an innovative and promising approach to support the urban design process. However, most deep generative urban form models are based on image representations that do not explicitly consider topological relationships among urban form elements. Toward developing an urban form generation framework aided by deep generative models and considering topological information, this paper reviews urban form generation, deep generative models/deep graph generation, and the state of the art of deep generative models in architectural and urban form generation. Based on the literature review, a topology-based urban form generation framework aided by deep generative models is proposed. The hypotheses of street network generation by deep generative models for graph generation and plot/building configuration generation by deep generative models/space syntax and the feasibility of the proposed framework require validation in future research. [ABSTRACT FROM AUTHOR]

Discovering hit molecules with desired biological activity in a directed manner is a promising but profound task in computer-aided drug discovery. Inspired by recent generative AI approaches, particularly Diffusion Models (DM), we propose Graph Latent Diffusion Model (GLDM)—a latent DM that preserves both the effectiveness of autoencoders of compressing complex chemical data and the DM's capabilities of generating novel molecules. Specifically, we first develop an autoencoder to encode the molecular data into low-dimensional latent representations and then train the DM on the latent space to generate molecules inducing targeted biological activity defined by gene expression profiles. Manipulating DM in the latent space rather than the input space avoids complicated operations to map molecule decomposition and reconstruction to diffusion processes, and thus improves training efficiency. Experiments show that GLDM not only achieves outstanding performances on molecular generation benchmarks, but also generates samples with optimal chemical properties and potentials to induce desired biological activity. [ABSTRACT FROM AUTHOR]

Local differential privacy algorithms combined with deep generative models can enhance secure medical image sharing among researchers in the public domain without central administrators; however, these images were limited to the generation of low-resolution images, which are very insufficient for diagnosis by medical doctors. To enhance the performance of deep generative models so that they can generate high-resolution medical images, we propose a large-scale diffusion model that can, for the first time, unconditionally generate high-resolution ( 256 × 256 × 256 ) volumetric medical images (head magnetic resonance images). This diffusion model has 19 billion parameters, but to make it easy to train it, we temporally divided the model into 200 submodels, each of which has 95 million parameters. Moreover, on the basis of this new diffusion model, we propose another formulation of image anonymization with which the processed images can satisfy provable Gaussian local differential privacy and with which we can generate images semantically different from the original image but belonging to the same class. We believe that the formulation of this new diffusion model and the implementation of local differential privacy algorithms combined with the diffusion models can contribute to the secure sharing of practical images upstream of data processing. [ABSTRACT FROM AUTHOR]

Large chemical spaces (CSs) include traditional large compound collections, combinatorial libraries covering billions to trillions of molecules, DNA-encoded chemical libraries comprising complete combinatorial CSs in a single mixture, and virtual CSs explored by generative models. The diverse nature of these types of CSs require different chemoinformatic approaches for navigation. An overview of different types of large CSs is provided. Molecular representations and similarity metrics suitable for large CS exploration are discussed. A summary of navigation of CSs in generative models is provided. Methods for characterizing and comparing CSs are discussed. The size of large CSs might restrict navigation to specialized algorithms and limit it to considering neighborhoods of structurally similar molecules. Efficient navigation of large CSs not only requires methods that scale with size but also requires smart approaches that focus on better but not necessarily larger molecule selections. Deep generative models aim to provide such approaches by implicitly learning features relevant for targeted biological properties. It is unclear whether these models can fulfill this ideal as validation is difficult as long as the covered CSs remain mainly virtual without experimental verification. [ABSTRACT FROM AUTHOR]

Water resources are essential for sustaining human life and promoting sustainable development. However, rapid urbanization and industrialization have resulted in a decline in freshwater availability. Effective prevention and control of water pollution are essential for ecological balance and human well-being. Water quality assessment is crucial for monitoring and managing water resources. Existing machine learning-based assessment methods tend to classify the results into the majority class, leading to inaccuracies in the outcomes due to the prevalent issue of imbalanced class sample distribution in practical scenarios. To tackle the issue, we propose a novel approach that utilizes the VAE–WGAN–GP model. The VAE–WGAN–GP model combines the encoding and decoding mechanisms of VAE with the adversarial learning of GAN. It generates synthetic samples that closely resemble real samples, effectively compensating data of the scarcity category in water quality evaluation. Our contributions include (1) introducing a deep generative model to alleviate the issue of imbalanced category samples in water quality assessment, (2) demonstrating the faster convergence speed and improved potential distribution learning ability of the proposed VAE–WGAN–GP model, (3) introducing the compensation degree concept and conducting comprehensive compensation experiments, resulting in a 9.7% increase in the accuracy of water quality assessment for multi-classification imbalance samples. HIGHLIGHTS Novel method: the VAE–WGAN–GP model is introduced to alleviate the problem of imbalanced category distribution in water quality evaluation and improve the accuracy of assessment.; Water resource management: our research bridges the gap in the distribution of categories in water management by providing deep generative models to compensate for data scarcity in water quality assessments.;

Abstract Differential analysis of bulk RNA-seq data often suffers from lack of good controls. Here, we present a generative model that replaces controls, trained solely on healthy tissues. The unsupervised model learns a low-dimensional representation and can identify the closest normal representation for a given disease sample. This enables control-free, single-sample differential expression analysis. In breast cancer, we demonstrate how our approach selects marker genes and outperforms a state-of-the-art method. Furthermore, significant genes identified by the model are enriched in driver genes across cancers. Our results show that the in silico closest normal provides a more favorable comparison than control samples.

Image Synthesis (IS), an expansion to Artificial Intelligence (AI) and Computer Vision, is the technique of artificially producing images that retains some specific required contents. An adequate procedure to handle IS problem is to tackle it using the Deep Generative Models. Generative Models are broadly utilized in numerous sub fields of AI and have empowered versatile demonstration of perplexing scenarios including image, text and music. In this paper, a particular class of Deep Generative model namely Generative Adversarial Networks (GAN) has been considered to provide a way to acquire deep illustrations derived from backpropagation signals and without the use of wide range of annotated training data. The design of GAN architecture plays a key role in image synthesis and the motive behind this paper is to analyse GAN architecture based on different variants of GANs with respect to Image Synthesis. Furthermore, a compact categorization of GANs along with their key features, pros and cons have been investigated to identify the research challenges in this field. [ABSTRACT FROM AUTHOR]

There are two main ways to discover or design small drug molecules. The first involves fine-tuning existing molecules or commercially successful drugs through quantitative structure-activity relationships and virtual screening. The second approach involves generating new molecules through de novo drug design or inverse quantitative structure-activity relationship. Both methods aim to get a drug molecule with the best pharmacokinetic and pharmacodynamic profiles. However, bringing a new drug to market is an expensive and timeconsuming endeavor, with the average cost being estimated at around $2.5 billion. One of the biggest challenges is screening the vast number of potential drug candidates to find one that is both safe and effective. The development of artificial intelligence in recent years has been phenomenal, ushering in a revolution in many fields. The field of pharmaceutical sciences has also significantly benefited from multiple applications of artificial intelligence, especially drug discovery projects. Artificial intelligence models are finding use in molecular property prediction, molecule generation, virtual screening, synthesis planning, repurposing, among others. Lately, generative artificial intelligence has gained popularity across domains for its ability to generate entirely new data, such as images, sentences, audios, videos, novel chemical molecules, etc. Generative artificial intelligence has also delivered promising results in drug discovery and development. This review article delves into the fundamentals and framework of various generative artificial intelligence models in the context of drug discovery via de novo drug design approach. Various basic and advanced models have been discussed, along with their recent applications. The review also explores recent examples and advances in the generative artificial intelligence approach, as well as the challenges and ongoing efforts to fully harness the potential of generative artificial intelligence in generating novel drug molecules in a faster and more affordable manner. Some clinical-level assets generated form generative artificial intelligence have also been discussed in this review to show the ever-increasing application of artificial intelligence in drug discovery through commercial partnerships. [ABSTRACT FROM AUTHOR]

Blind image deblurring (BID) has been extensively studied in computer vision and adjacent fields. Modern methods for BID can be grouped into two categories: single-instance methods that deal with individual instances using statistical inference and numerical optimization, and data-driven methods that train deep-learning models to deblur future instances directly. Data-driven methods can be free from the difficulty in deriving accurate blur models, but are fundamentally limited by the diversity and quality of the training data—collecting sufficiently expressive and realistic training data is a standing challenge. In this paper, we focus on single-instance methods that remain competitive and indispensable. However, most such methods do not prescribe how to deal with unknown kernel size and substantial noise, precluding practical deployment. Indeed, we show that several state-of-the-art (SOTA) single-instance methods are unstable when the kernel size is overspecified, and/or the noise level is high. On the positive side, we propose a practical BID method that is stable against both, the first of its kind. Our method builds on the recent ideas of solving inverse problems by integrating physical models and structured deep neural networks, without extra training data. We introduce several crucial modifications to achieve the desired stability. Extensive empirical tests on standard synthetic datasets, as well as real-world NTIRE2020 and RealBlur datasets, show the superior effectiveness and practicality of our BID method compared to SOTA single-instance as well as data-driven methods. The code of our method is available at https://github.com/sun-umn/Blind-Image-Deblurring. [ABSTRACT FROM AUTHOR]

This article proposes a sensor data anonymization model that is trained on decentralized data and strikes a desirable trade-off between data utility and privacy, even in heterogeneous settings where the sensor data have different underlying distributions. Our anonymization model, dubbed Blinder, is based on a variational autoencoder and one or multiple discriminator networks trained in an adversarial fashion.We use the modelagnostic meta-learning framework to adapt the anonymization model trained via federated learning to each user's data distribution. We evaluate Blinder under different settings and show that it provides end-to-end privacy protection on two IMU datasets at the cost of increasing privacy loss by up to 4.00% and decreasing data utility by up to 4.24%, compared to the state-of-the-art anonymization model trained on centralized data. We also showcase Blinder's ability to anonymize the radio frequency sensing modality. Our experiments con- firm that Blinder can obscure multiple private attributes at once, and has sufficiently low power consumption and computational overhead for it to be deployed on edge devices and smartphones to perform real-time anonymization of sensor data. [ABSTRACT FROM AUTHOR]