There has been an increasing interest in utilizing machine learning methods in inverse problems and imaging. Most of the work has, however, concentrated on image reconstruction problems, and the number of studies rega...
详细信息
There has been an increasing interest in utilizing machine learning methods in inverse problems and imaging. Most of the work has, however, concentrated on image reconstruction problems, and the number of studies regarding the full solution of the inverse problem is limited. In this work, we study a machine learning--based approach for the Bayesian inverse problem of photoacoustic tomography. We develop an approach for estimating the posterior distribution in photoacoustic tomography using an approach based on the variational autoencoder. The approach is evaluated with numerical simulations and compared to the solution of the inverse problem using a Bayesian approach.
Neuroimaging-derived brain age has been identified as a promising biomarker for accelerated brain age;however, the ageing process is highly heterogeneous and there is a need to further study the different brain ageing...
详细信息
ISBN:
(纸本)9783031745607;9783031745614
Neuroimaging-derived brain age has been identified as a promising biomarker for accelerated brain age;however, the ageing process is highly heterogeneous and there is a need to further study the different brain ageing trajectories. In this study, we implemented a variational autoencoder (VAE) based model coupled with regression to identify different age-related patterns. Additionally, we correlated the patterns obtained, using a linear regression approach, with dementia-related risk factors. The model was evaluated in different cohorts, UK Biobank and ALFA+, to assess the robustness of the approach. The results showed a feasible strategy for detecting and validating brain age-related trajectories to identify possible early deviations using morphological brain data.
We propose a hybrid method for generating arbitrage-free implied volatility (IV) surfaces consistent with historical data by combining model-free variational autoencoders (VAEs) with continuous time stochastic differe...
详细信息
We propose a hybrid method for generating arbitrage-free implied volatility (IV) surfaces consistent with historical data by combining model-free variational autoencoders (VAEs) with continuous time stochastic differential equation (SDE) driven models. We focus on two classes of SDE models: regime switching models and Le'\vy additive processes. By projecting historical surfaces onto the space of SDE model parameters, we obtain a distribution on the parameter subspace faithful to the data on which we then train a VAE. Arbitrage-free IV surfaces are then generated by sampling from the posterior distribution on the latent space, decoding to obtain SDE model parameters, and finally mapping those parameters to IV surfaces. We further refine the VAE model by including conditional features and demonstrate its superior generative out-of-sample performance. Finally, we showcase how our method can be used as a data augmentation tool to help practitioners manage the tail risk of option portfolios.
In this paper, we present Period Singer, a novel end-to-end singing voice synthesis (SVS) model that utilizes variational inference for periodic and aperiodic components, aimed at producing natural-sounding waveforms....
详细信息
In this paper, we present Period Singer, a novel end-to-end singing voice synthesis (SVS) model that utilizes variational inference for periodic and aperiodic components, aimed at producing natural-sounding waveforms. Recent end-to-end SVS models have demonstrated the capability of synthesizing high-fidelity singing voices. However, owing to deterministic pitch conditioning, they do not fully address the one-to-many problem. To address this problem, we present the Period Singer architecture, which integrates variational autoencoders for the periodic and aperiodic components. Additionally, our methodology eliminates the dependency on an external aligner by estimating the phoneme alignment through a monotonic alignment search within note boundaries. Our empirical evaluations show that Period Singer outperforms existing end-to-end SVS models on Mandarin and Korean datasets. The efficacy of the proposed method was further corroborated by ablation studies.
We present the new bidirectional variational autoencoder (BVAE) network architecture. The BVAE uses a single neural network both to encode and decode instead of an encoder-decoder network pair. The network encodes in ...
详细信息
ISBN:
(纸本)9798350359329;9798350359312
We present the new bidirectional variational autoencoder (BVAE) network architecture. The BVAE uses a single neural network both to encode and decode instead of an encoder-decoder network pair. The network encodes in the forward direction and decodes in the backward direction through the same synaptic web. Simulations compared BVAEs and ordinary VAEs on the four image tasks of image reconstruction, classification, interpolation, and generation. The image datasets included MNIST handwritten digits, Fashion-MNIST, CIFAR-10, and CelebA-64 face images. The bidirectional structure of BVAEs cut the parameter count by almost 50% and still slightly outperformed the unidirectional VAEs.
Brain computer interfaces based on speech imagery have attracted attention in recent years as more flexible tools of machine control and communication. Classifiers of imagined speech are often trained for each individ...
详细信息
ISBN:
(纸本)9789464593617;9798331519773
Brain computer interfaces based on speech imagery have attracted attention in recent years as more flexible tools of machine control and communication. Classifiers of imagined speech are often trained for each individual due to individual differences in brain activity. However, the amount of brain activity data that can be measured from a single person is often limited, making it difficult to train a model with high classification accuracy. In this study, to improve the performance of the classifiers for each individual, we trained variational autoencoders (VAEs) using magnetoencephalographic (MEG) data from seven participants during speech imagery. The trained encoders of VAEs were transferred to EEGNet, which classified speech imagery MEG data from another participant. We also trained conditional VAEs to augment the training data for the classifiers. The results showed that the transfer learning improved the performance of the classifiers for some participants. Data augmentation also improved the performance of the classifiers for most participants. These results indicate that the use of VAE feature representations learned using MEG data from multiple individuals can improve the classification accuracy of imagined speech from a new individual even when a limited amount of MEG data is available from the new individual.
In this work, we propose to utilize a variational autoencoder (VAE) for channel estimation (CE) in underdetermined (UD) systems. The basis of the method forms a recently proposed concept in which a VAE is trained on c...
详细信息
ISBN:
(纸本)9798350344868;9798350344851
In this work, we propose to utilize a variational autoencoder (VAE) for channel estimation (CE) in underdetermined (UD) systems. The basis of the method forms a recently proposed concept in which a VAE is trained on channel state information (CSI) data and used to parameterize an approximation to the mean squared error (MSE)-optimal estimator. The contributions in this work extend the existing framework from fully-determined (FD) to UD systems, which are of high practical relevance. Particularly noteworthy is the extension of the estimator variant, which does not require perfect CSI during its offline training phase. This is a significant advantage compared to most other deep learning (DL)-based CE methods, where perfect CSI during the training phase is a crucial prerequisite. Numerical simulations for hybrid and wideband systems demonstrate the excellent performance of the proposed methods compared to related estimators.
Collision Risk Models (CRM), used by regulatory authorities to approve new procedures and/or technology, assess the probability of air-to-air collisions against a Target Level of Safety (e.g. 10E-9). A key component o...
详细信息
ISBN:
(纸本)9798350393101;9798350393095
Collision Risk Models (CRM), used by regulatory authorities to approve new procedures and/or technology, assess the probability of air-to-air collisions against a Target Level of Safety (e.g. 10E-9). A key component of the CRM is a variety of plausible flight tracks. Historic flight tracks can be blended with simulated flight tracks to generate a sample of tracks for analysis, however, even using super-computers, the sample size of flight tracks is not sufficient for a narrow confidence interval for the probability of a collision. This paper describes the results of the application of an AI/ML algorithm known as variational autoencoders (VAE) to increase the sample size of flight tracks for CRM. 2,356 flight tracks for the Ondre 1 arrival to Runway 26R at Atlanta airport were used to train a VAE. Next a single "seed" track, representing a nominal trajectory, was used to generate 2,356 synthetic flight tracks. By properly adjusting the VAE coefficients, the generated synthetic tracks matched historic tracks with regards to mean Along Track Distance (ATD), Total Cross Track Distance (TCTD), vertical dispersion at 9nm and 6nm from the Runway, and lateral dispersion at 9nm and 6nm from the Runway. However, in all the metrics the variability in the flight tracks was significantly smaller than the variability in the historic tracks. These promising results suggest that an approach using multiple "seed tracks" could achieve the desired variability. The implications of these results and future work are discussed.
Albeit of crucial interest for financial researchers, market-implied volatility data of European swaptions often exhibit large portions of missing quotes due to illiquidity of the underlying swaption instruments. In t...
详细信息
作者：
Ma, HeHarbin Engn Univ
Coll Intelligent Syst Sci & Engn Harbin 150000 Peoples R China
Clustering is an important and challenging research topic in many fields. Although various clustering algorithms have been developed in the past, traditional shallow clustering algorithms cannot mine the underlying st...
详细信息
Clustering is an important and challenging research topic in many fields. Although various clustering algorithms have been developed in the past, traditional shallow clustering algorithms cannot mine the underlying structural information of the data. Recent advances have shown that deep clustering can achieve excellent performance on clustering tasks. In this work, a novel variational autoencoder-based deep clustering algorithm is proposed. It treats the Gaussian mixture model as the prior latent space and uses an additional classifier to distinguish different clusters in the latent space accurately. A similarity-based loss function is proposed consisting specifically of the cross-entropy of the predicted transition probabilities of clusters and the Wasserstein distance of the predicted posterior distributions. The new loss encourages the model to learn meaningful cluster-oriented representations to facilitate clustering tasks. The experimental results show that our method consistently achieves competitive results on various data sets.
暂无评论