State-of-the-art recommender systems (RSs) generally try to improve the overall recommendation quality. However, users usually tend to explicitly filter the item set based on available categories, e.g., smartphone bra...
详细信息
ISBN:
(纸本)9783031159312;9783031159305
State-of-the-art recommender systems (RSs) generally try to improve the overall recommendation quality. However, users usually tend to explicitly filter the item set based on available categories, e.g., smartphone brands, movie genres. For this reason, an RS that can make this step automatically is likely to increase the user's experience. This paper proposes a Conditioned variational autoencoder (C-VAE) for constrained top-N item recommendation where the recommended items must satisfy a given condition. The proposed model architecture is similar to a standard VAE in which a condition vector is fed into the encoder. The constrained ranking is learned during training thanks to a new reconstruction loss that takes the input condition into account. We show that our model generalizes the state-of-the-art Mult-VAE collaborative filtering model. Experimental results underline the potential of CVAE in providing accurate recommendations under constraints. Finally, the performed analyses suggest that C-VAE can be used in other recommendation scenarios, such as context-aware recommendation.
Out-of-distribution (OOD) detection, i.e., finding test samples derived from a different distribution than the training set, as well as reasoning about such samples (OOD reasoning), are necessary to ensure the safety ...
详细信息
ISBN:
(纸本)9781665470926
Out-of-distribution (OOD) detection, i.e., finding test samples derived from a different distribution than the training set, as well as reasoning about such samples (OOD reasoning), are necessary to ensure the safety of results generated by machine learning models. Recently there have been promising results for OOD detection in the latent space of variational autoencoders (VAEs). However, without disentanglement, VAEs cannot perform OOD reasoning. Disentanglement ensures a one-to-many mapping between generative factors of OOD (e.g., rain in image data) and the latent variables to which they are encoded. Although previous literature has focused on weakly-supervised disentanglement on simple datasets with known and independent generative factors. In practice, achieving full disentanglement through weak supervision is impossible for complex datasets, such as Carla, with unknown and abstract generative factors. As a result, we propose an OOD reasoning framework that learns a partially disentangled VAE to reason about complex datasets. Our framework consists of three steps: partitioning data based on observed generative factors, training a VAE as a logic tensor network that satisfies disentanglement rules, and run-time OOD reasoning. We evaluate our approach on the Carla dataset and compare the results against three state-of-the-art methods. We found that our framework outperformed these methods in terms of disentanglement and end-to-end OOD reasoning.
Deep learning has gotten much attention in industrial field, many fault detection methods based on deep learning have been developed for nonlinear industrial processes. However, most of them do not take the quality-re...
详细信息
Deep learning has gotten much attention in industrial field, many fault detection methods based on deep learning have been developed for nonlinear industrial processes. However, most of them do not take the quality-related faults into account. In order to extract the latent variables which can represent the separated quality-related and unrelated information, this paper proposes a novel deep VIB-VAE algorithm, which combines variational autoencoder (VAE) model and deep variational information bottleneck (VIB). Deep VIB extracts quality-related latent variables by maximizing mutual information between latent variables and process quality while minimizing mutual information between latent variables and observation. VAE is used to learn the quality-unrelated part with above quality-related latent variables as auxiliary information. To monitor and distinguish quality-related and quality-unrelated faults, two monitoring statistics are designed by the two-part latent variables. The reconstruction error by VAE is introduced to improve the performance of fault detection. Finally, the effectiveness of the proposed deep VIB-VAE algorithm is demonstrated by a numerical case and a real hot strip mill process case, respectively. (C) 2021 ISA. Published by Elsevier Ltd. All rights reserved.
Fire detection is a critical component of a building safety monitoring system and remains an important research area with weighty practical relevance. Significant advances have occurred in recent years in building aut...
详细信息
Fire detection is a critical component of a building safety monitoring system and remains an important research area with weighty practical relevance. Significant advances have occurred in recent years in building automation, and the operation of buildings has become more complex and requires ever more effective monitoring systems. In this work, we develop a novel fire detection method using deep Long-Short Term Memory (LSTM) neural networks and variational autoencoder (VAE) to meet these increasingly stringent requirements and outperform existing fire detection methods. To evaluate the effectiveness of our method, we develop high-fidelity simulations, and we use datasets from real-world fire and non-fire experiments provided by NIST. We compare and discuss the performance of our proposed fire detection with alternative methods, including the standard LSTM, cumulative sum control chart (CUSUM), exponentially weighted moving average (EWMA), and two currently used fixed-temperature heat detectors. The results using the simulation-based and the real-world experiments are complementary, and they indicate that the LSTM-VAE robustly outperforms the other detection methods with, for example, statistically significant shorter alarm time lags, no missed detection, and no false alarms. The results also identify shortcomings of other detection methods and indicate a clear ranking among them (LSTM-VAE > EWMA > LSTM > CUSUM).
In present study, we proposed a general framework based on a convolutional kernel and a variational autoencoder (CVAE) for anomaly detection on both complex image and vector datasets. The main idea is to maximize mutu...
详细信息
In present study, we proposed a general framework based on a convolutional kernel and a variational autoencoder (CVAE) for anomaly detection on both complex image and vector datasets. The main idea is to maximize mutual information (MMI) through regularizing key information as follows: (1) the features between original input and the representation of latent space, (2) that between the first convolutional layer output and the last convolutional layer input, (3) original input and output of the decoder to train the model. Therefore, the proposed CVAE is optimized by combining the representations learned across the three different objectives targeted at MMI on both local and global variables with the original training objective function of Kullback-Leibler divergence distributions. It allowed achieving the additional supervision power for the detection of image and vector data anomalies using convolutional and fully connected layers, respectively. Our proposal CVAE combined by regularizing multiple discriminator spaces to detect anomalies was introduced for the first time as far as we know. To evaluate the reliability of the proposed CVAE-MMI, it was compared with the convolutional autoencoder-based model using the original objective function. Furthermore, the performance of our network was compared over state-of-the-art approaches in distinguishing anomalies concerning both image and vector datasets. The proposed structure outperformed the state-of-the-arts with high and stable area under the curve values.
At present times, COVID-19 has become a global illness and infected people has increased exponentially and it is difficult to control due to the non-availability of large quantity of testing kits. Artificial intellige...
详细信息
At present times, COVID-19 has become a global illness and infected people has increased exponentially and it is difficult to control due to the non-availability of large quantity of testing kits. Artificial intelligence (AI) techniques including machine learning (ML), deep learning (DL), and computer vision (CV) approaches find useful for the recognition, analysis, and prediction of COVID-19. Several ML and DL techniques are trained to resolve the supervised learning issue. At the same time, the potential measure of the unsupervised learning technique is quite high. Therefore, unsupervised learning techniques can be designed in the existing DL models for proficient COVID-19 prediction. In this view, this paper introduces a novel unsupervised DL based variational autoencoder (UDL-VAE) model for COVID-19 detection and classification. The UDL-VAE model involved adaptive Wiener filtering (AWF) based preprocessing technique to enhance the image quality. Besides, Inception v4 with Adagrad technique is employed as a feature extractor and unsupervised VAE model is applied for the classification process. In order to verify the superior diagnostic performance of the UDL-VAE model, a set of experimentation was carried out to highlight the effective outcome of the UDL-VAE model. The obtained experimental values showcased the effectual results of the UDL-VAE model with the higher accuracy of 0.987 and 0.992 on the binary and multiple classes respectively. (c) 2021 Elsevier B.V. All rights reserved.
The embedding representation of the case text represent text as vector which consist information of original texts abundantly. Text embedding representation usually uses text statistical features or content features a...
详细信息
The embedding representation of the case text represent text as vector which consist information of original texts abundantly. Text embedding representation usually uses text statistical features or content features alone. However, case texts have characteristics that include similar structure, repeated words, and different text lengths. And the statistical feature or content feature cannot represent case text efficiently. In this paper, we propose a joint variational autoencoder (VAE) to represent case text embedding representation. We consider the statistical features and content features of case texts together, and use VAE to align the two features into the same space. We compare our representations with existing methods in terms of quality, relationship, and efficiency. The experiment results show that our method has achieved good results, which have higher performance than the model using single feature.
Arrhythmia has become one of the important causes of human death. The research on arrhythmia detection has great medical value. In reality, patients' arrhythmia heartbeat is much less than the normal heartbeat. Su...
详细信息
ISBN:
(纸本)9783031155123;9783031155116
Arrhythmia has become one of the important causes of human death. The research on arrhythmia detection has great medical value. In reality, patients' arrhythmia heartbeat is much less than the normal heartbeat. Supervised classifiers often have the problem of imbalanced training data. Therefore, we propose an unsupervised personalized arrhythmia detection system, called PerAD. PerAD trains a lightweight autoencoder ShaAE for each user for arrhythmia detection. ShaAE only needs to use the user's personal normal data for training. The encoder and decoder of ShaAE are composed of a lightweight network ShaRNN. ShaRNN is a two-layer RNN structure that can process data in parallel. Thus, ShaAE is easy to deploy to edge wearable devices. We also design a fast-inference variational autoencoder to generate normal simulation samples to assist in training ShaAE. We test ShaAE on MIT-BIH Arrhythmia Database. ShaAE without using simulation data to assist training can achieve 96.86% accuracy. ShaAE using simulation samples to assist training can achieve accuracy of 97.11% and has 6.19% higher performance than state-of-the-art for f1 score.
Cross-domain recommendation (CDR) leverages knowledge from the source domain to make recommendations for the cold-start users in the target domain. On account of fully utilizing information, various relationships amon...
详细信息
Cross-domain recommendation (CDR) leverages knowledge from the source domain to make recommendations for the cold-start users in the target domain. On account of fully utilizing information, various relationships among users and items are taken into account, i.e., the interaction relationship between users and their corresponding items;the relationship among users or items;and the indirect relationship between the user and items related to other users. In order to process these relationships, we propose a novel framework named Memory Pool variational autoencoder (MPVAE). The main advantages of the MPVAE model lie in three aspects: (1) it generates the embedding representations that incorporate more information by a memory pool mechanism in the source and target domains;(2) it involves the relationship among users or items efficiently by the similarity measurement, further, the indirect relationship can be explicitly described, which makes full use of information in the source domain;and (3) it leverages the superiority of the probability model from the perspective of the VAE structure, which ensures generation and robustness. Comprehensive experiments on three real datasets show that the proposed model achieves remarkable superiority over several competitive CDR models.
This paper proposes an autoregressive speech synthesis model based on the variational autoencoder incorporating latent sequence representation for acoustic and linguistic features and the structure of a hidden semi-Ma...
详细信息
ISBN:
(纸本)9781665405409
This paper proposes an autoregressive speech synthesis model based on the variational autoencoder incorporating latent sequence representation for acoustic and linguistic features and the structure of a hidden semi-Markov model (HSMM). Although autoregressive models can provide efficient and accurate modeling of acoustic features, they have exposure bias, i.e., the mismatch between training (teacher-forcing) and inference (free-running). To overcome this problem, we introduce an autoregressive latent variable sequence, rather than using autoregressive generation of observations. Latent representation of alignment using HSMM-based structured attention mechanism enables the use of a completely consistent training algorithm for acoustic modeling with explicit duration models. Experimental results indicate that the proposed model outperformed baselines in subjective naturalness.
暂无评论