Without knowledge of specific pockets, generating ligands based on the global structure of a protein target plays a crucial role in drug discovery as it helps reduce the search space for potential drug-like candidates...
详细信息
Without knowledge of specific pockets, generating ligands based on the global structure of a protein target plays a crucial role in drug discovery as it helps reduce the search space for potential drug-like candidates in the pipeline. However, contemporary methods require optimizing tailored networks for each protein, which is arduous and costly. To address this issue, we introduce TargetVAE, a target-aware variational auto-encoder that generates ligands with desirable properties including high binding affinity and high synthesizability to arbitrary target proteins, guided by a multimodal deep neural network built based on geometric and sequence models, named Protein Multimodal Network (PMN), as the prior for the generative model. PMN unifies different representations of proteins (e.g. primary structure-sequence of amino acids, 3D tertiary structure, and residue-level graph) into a single representation. Our multimodal architecture learns from the entire protein structure and is able to capture their sequential, topological, and geometrical information by utilizing language modeling, graph neural networks, and geometric deep learning. We showcase the superiority of our approach by conducting extensive experiments and evaluations, including predicting protein-ligand binding affinity in the PBDBind v2020 dataset as well as the assessment of generative model quality, ligand generation for unseen targets, and docking score computation. Empirical results demonstrate the promising and competitive performance of our proposed approach. Our software package is publicly available at https://***/HySonLab/Ligand_Generation.
作者:
Zhang, YechuanZheng, Jian-QingChappell, MichaelUniv Oxford
Inst Biomed Engn Dept Engn Sci Oxford OX1 3PJ England Univ Oxford
Kennedy Inst Rheumatol Nuffield Dept Orthopaed Rheumatol & Musculoskeleta Oxford OX3 7FY England Univ Nottingham
Sir Peter Mansfield Imaging Ctr Sch Med Nottingham NG7 2RD England Univ Nottingham
Sch Med Mental Hlth & Clin Neurosci Nottingham NG7 2RD England Univ Oxford
Wellcome Ctr Integrat Neuroimaging Nuffield Dept Clin Neurosci FMRIB Oxford OX3 9DU England
In this paper, a variational autoencoder (VAE) based framework is introduced to solve parameter estimation problems for non-linear forward models. In particular, we focus on applications in the field of medical imagin...
详细信息
In this paper, a variational autoencoder (VAE) based framework is introduced to solve parameter estimation problems for non-linear forward models. In particular, we focus on applications in the field of medical imaging where many thousands of model-based inference analyses might be required to populate a single parametric map. We adopt the concept from variational Bayes (VB) of using an approximate representation of the posterior, and the concept from the VAE of using the latent space representation to encode the parameters of a forward model. Our work develops the idea of mapping between time-series data and latent parameters using a neural network in variational way. A loss function that differs from the classic VAE formulation and a new sampling strategy are proposed to enable uncertainty estimation as part of the forward model inference. The VAE-based structure is evaluated using simulation experiments on a simple example and two perfusion MRI forward models. Compared with analytical VB (aVB) and Markov Chain Monte Carlo (MCMC), our VAE-based model achieves comparable accuracy, and hundredfold improvement in computational time (100ms/image). We believe this VAE-like framework can be generalized to imaging modularities with higher complexity and thus benefit clinical adoption where otherwise long processing time associated with conventional inference methods is prohibitive.
Medical image segmentation can provide a reliable basis for clinical analysis and diagnosis. However, this task is challenging due to the low contrast, boundary ambiguity between organs or lesions and surrounding tiss...
详细信息
Medical image segmentation can provide a reliable basis for clinical analysis and diagnosis. However, this task is challenging due to the low contrast, boundary ambiguity between organs or lesions and surrounding tissues, and noise interference of images. To address this challenge, which is unique to medical images, and further improve the segmentation accuracy and precision, a medical image segmentation model (TransDiff) is proposed from the perspective of improving model robustness and enriching semantic information. TransDiff comprises three parts: a variational autoencoder (VAE), a diffusion transformer model and a Swin Transformer. The VAE constructs a latent space to provide an environment for fully extracting and fusing features. The diffusion model predicts and removes noise by inferring semantics through the propagation of information between nodes. The Swin Transformer enriches discriminative features as a conditional part. TransDiff inherits the robustness to noise and missing data of the diffusion model and the feature enrichment of the Swin Transformer, thus exhibiting a higher understanding of semantic information. It performs well on medical datasets with three different image modalities, outperforms existing medical image segmentation methods in terms of segmentation precision and accuracy, and has good generalizability. The codes and trained models will be publicly available at https://***/xiaoxiao1997/TransDiff.
Efficient channel state information (CSI) compression and feedback from user equipment to the base station (BS) are crucial for achieving the promised capacity gains in massive multiple-input multiple-output (MIMO) sy...
详细信息
Efficient channel state information (CSI) compression and feedback from user equipment to the base station (BS) are crucial for achieving the promised capacity gains in massive multiple-input multiple-output (MIMO) systems. Deep autoencoder (AE)-based schemes have been proposed to improve the efficiency of CSI compression and feedback. However, existing AE-based schemes suffer from critical issues in both CSI dimensionality reduction and latent feature quantization. In this paper, we propose a novel hierarchical sparse AE for efficient CSI compression and feedback for the 5G-NR fixed-length CSI feedback mechanism. Our approach employs a two-tier AE structure to jointly compress the sparse CSI latent feature and its side information. Additionally, we utilize a model-assisted Bayesian Rate-Distortion approach to train the weights of the AE. Specifically, the training loss function is formulated based on the variational Bayesian inference framework given a parametric Bernoulli Laplace Mixture prior model and a sparsity-inducing likelihood model. Furthermore, we propose a model-assisted adaptive coding algorithm to quantize the latent feature under the fixed codeword bit length constraint. Our experimental results demonstrate that the proposed solution outperforms existing AE-based schemes under various feedback budgets.
Detecting anomalous executions in business process data is crucial for safeguarding the efficiency and success of an organization. Unsupervised approaches are commonly used for business process anomaly detection becau...
详细信息
Detecting anomalous executions in business process data is crucial for safeguarding the efficiency and success of an organization. Unsupervised approaches are commonly used for business process anomaly detection because of the scarcity of labeled anomaly data. However, these approaches often encounter a notable decline in performance because they lack prior knowledge about the anomalies. Additionally, most of them do not perform root cause analysis on the detected anomalies. This study proposes a variational autoencoder-based approach to overcome the performance limitations of existing unsupervised methods and determine the root causes of the detected anomalies. The learning of the variational autoencoder from unlabeled business process data is enhanced in the proposed approach by leveraging different architectural components, namely, the entity embedding technique, the bidirectional long short-term memory network, and the self-attention mechanism. Combining these architectural components in the variational autoencoder architecture leads to learning high-level representations from the business process data and thus improving the reconstruction capability of the variational autoencoder. Furthermore, this study suggests feeding the reconstruction error provided by the variational autoencoder into the logistic regression classifier to improve the accuracy of anomaly detection. The performance of the proposed model was evaluated on real-life and synthetic datasets. The experimental findings indicate that the proposed model outperforms six existing anomaly detection models in terms of precision, recall, and F1-score metrics.
To address the class imbalance issue in network intrusion detection, which degrades performance of intrusion detection models, this paper proposes a novel generative model called VAE-WACGAN to generate minority class ...
详细信息
To address the class imbalance issue in network intrusion detection, which degrades performance of intrusion detection models, this paper proposes a novel generative model called VAE-WACGAN to generate minority class samples and balance the dataset. This model extends the variational autoencoder Generative Adversarial Network (VAEGAN) by integrating key features from the Auxiliary Classifier Generative Adversarial Network (ACGAN) and the Wasserstein Generative Adversarial Network with Gradient Penalty (WGAN-GP). These enhancements significantly improve both the quality of generated samples and the stability of the training process. By utilizing the VAE-WACGAN model to oversample anomalous data, more realistic synthetic anomalies that closely mirror the actual network traffic distribution can be generated. This approach effectively balances the network traffic dataset and enhances the overall performance of the intrusion detection model. Experimental validation was conducted using two widely utilized intrusion detection datasets, UNSW-NB15 and CIC-IDS2017. The results demonstrate that the VAE-WACGAN method effectively enhances the performance metrics of the intrusion detection model. Furthermore, the VAE-WACGAN-based intrusion detection approach surpasses several other advanced methods, underscoring its effectiveness in tackling network security challenges.
Identification of intra-pulse modulation (IPM) of radar signals is a crucial part of contemporary electronic support systems and electronic intelligence reconnaissance. Artificial intelligence (AI)-based methods can b...
详细信息
Identification of intra-pulse modulation (IPM) of radar signals is a crucial part of contemporary electronic support systems and electronic intelligence reconnaissance. Artificial intelligence (AI)-based methods can be very effective in recognising the IPM of radar signals. In this direction, an automatic method is proposed for recognising a few IPMs of radar signals based on continuous wavelet transform (CWT) and a hybrid model of self-attention (SA)-aided convolutional neural network (CNN) and bidirectional long short-term memory (BiLSTM). Firstly, time-frequency attributes of different radar signals are obtained using CWT, and thereafter CNN-SA-BiLSTM is utilised for feature extraction from the 2D scalograms formed by the time-frequency components. The CNN extracts features from the scalograms, SA enhances the discriminative power of the feature map, and BiLSTM detects radar signals based on these features. Additionally, the study addresses real-world data imbalance issues by incorporating a generative AI model, namely the variational autoencoder (VAE). The VAE-based approach effectively mitigates challenges arising from data imbalance situations. This method is tested at varying noise levels to give a proper representation of the actual electronic warfare environment. The simulation results demonstrate that the best overall recognition accuracy of the proposed method is 98.4%, even at low signal-to-noise ratios (SNR).
With the rapid expansion of the vehicular cybersecurity (VCS) market and the increasing sophistication of cyberthreats, developing an adaptive intra-vehicular intrusion detection system (IDS) is crucial. This paper in...
详细信息
With the rapid expansion of the vehicular cybersecurity (VCS) market and the increasing sophistication of cyberthreats, developing an adaptive intra-vehicular intrusion detection system (IDS) is crucial. This paper introduces GenCoder, a generative artificial intelligence (GenAI)-based IDS that uniquely addresses the dynamic and evolving nature of vehicular cyberthreats. GenCoder combines a five-layer deep neural network (DNN) with a variational autoencoder (VAE), overseen by a novel communication layer known as the GenCoder layer. This system dynamically adapts to new intrusion patterns by generating and utilizing new training data when deviations from known patterns are detected. The generated samples have a Shannon entropy (SE) value of 1.65 bits for four classes, indicating standard variety among the synthetic data. GenCoder demonstrates exceptional adaptability, pushing the accuracy, precision, recall, and F1-score from 84.79%, 83.58%, 83.70%, and 83.64% to 92.19%, 90.12%, 90.44%, and 90.28%, respectively, after introducing 50% feature deformation to testing data. The novel concept of an adaptive intra-vehicular IDS, the innovative GenCoder layer that establishes seamless communication among the DNN, the VAE, and the dataset, as well as the unique assessment strategies of adaptability make this research exceptional with the potential to create a new dimension in automotive IDS research.
The wear modeling and life prediction of pantographs are crucial for ensuring the safety and reliability of urban rail transit systems. However, because of the complex interplay between stochastic vibrations and elect...
详细信息
The wear modeling and life prediction of pantographs are crucial for ensuring the safety and reliability of urban rail transit systems. However, because of the complex interplay between stochastic vibrations and electrical currents, pantograph wear exhibits strong variability, and physics-based degradation predictions based solely on material parameters, environmental factors, and wear mechanisms are limited in accuracy. Purely data-driven approaches, on the other hand, are constrained by their reliance on large datasets and lack of interpretability, making them difficult to meet practical engineering needs. To address these challenges, we propose an interpretable variational model called IV-NBEATS. This study integrates the surface wear mechanism under the asperity hypothesis into the N-BEATS model using the projection principle, thereby enhancing the interpretability of the model. In addition, we introduce a method for describing the uncertainty of key wear parameters, enabling a deep network to represent the uncertainty of these parameters. Furthermore, to cope with dynamic changes in wear system parameters, we propose a dynamic updating method based on a Bayesian directed graph model that effectively overcomes the limitations of existing methods in capturing the temporal evolution of wear system parameters. Finally, the effectiveness of the proposed approach is demonstrated through the analysis of a real-world case study of pantograph wear.
Denoising Diffusion Probabilistic Models (DDPMs) have demonstrated remarkable results in image generation. However, there exist a mismatch between the training and sampling process in current diffusion models, in addi...
详细信息
Denoising Diffusion Probabilistic Models (DDPMs) have demonstrated remarkable results in image generation. However, there exist a mismatch between the training and sampling process in current diffusion models, in addition, the U-Net denoising network based on simple residual blocks cannot predict noise information accurately, which affects the generated quality. To address these limitations, we present a novel image generation method that achieves higher fidelity. First, by additionally adding the standard Gaussian noise in the diffusion forward process, which does not disrupt the forward process, our method alleviates the mismatch. Subsequently, an important efficient denoising network based on U-Net is presented, where our proposed Simple Squeeze-Excitation and Simple GLU, combined with Depthwise Separable Convolution, enhance the ability of the model to predict real noise using the Simplified Nonlinear No Activation (SNNA) block. Furthermore, considering the structural characteristics of the baseline model, we introduce an additional cross-attention mechanism to enable DDPM to focus on VAE stage characteristics. Allowing the model to more accurately capture and learn the noise information. Finally, it is shown after extensive experiments the proposed DiffuseVAE++ obtains significant gains in FID scores, improving from 3.84 to 2.41 on CIFAR-10 and from 3.94 to 2.30 on CelebA-64. In particular, the IS scores on CIFAR-10 reaches 10.10, which is comparable to the current state-of-the-art methods competitively (e.g., U-ViT, StyleGAN2).
暂无评论