•This article formulates the Air Traffic Controllers’ (ATCOs’) reaction problem;•proposes a data-driven method simulating the uncertainty in the trajectories’ evolution;•proposes a methodology for ...evaluating methods resolving the ATCOs’ reaction problem;•proposes using a Variational Auto-Encoder to model ATCOs’ reactions;•evaluates the proposed method using real world data, also w.r.t. a baseline method.
With the aim to enhance automation in conflict detection and resolution (CD&R) tasks in the Air Traffic Management domain, this article proposes deep learning (DL) techniques that model Air Traffic Controllers’ reactions in resolving conflicts violating aircraft trajectories separation minimum constraints: This implies learning when the Air Traffic Controller reacts towards resolving a conflict, and how he/she reacts. Timely reactions, to which this article aims, focus on when do reactions happen, aiming to predict the trajectory points, as the aircraft state evolves, that the Air Traffic Controller (ATCO) issues a conflict resolution action. Towards this goal, the article formulates the Air Traffic Controllers’ reaction prediction problem for CD&R, presents DL methods that can model Air Traffic Controllers’ timely reactions, and evaluates these methods in real-world data sets, showing their efficacy in solving the problem with very high accuracy.
The increasing availability of structured but high dimensional data has opened new opportunities for optimization. One emerging and promising avenue is the exploration of unsupervised methods for ...projecting structured high dimensional data into low dimensional continuous representations, simplifying the optimization problem and enabling the application of traditional optimization methods. However, this line of research has been purely methodological with little connection to the needs of practitioners so far. In this article, we study the effect of different search space design choices for performing Bayesian optimization in high dimensional structured datasets. In particular, we analyses the influence of the dimensionality of the latent space, the role of the acquisition function and evaluate new methods to automatically define the optimization bounds in the latent space. Finally, based on experimental results using synthetic and real datasets, we provide recommendations for the practitioners.
In recent years, research on image generation has been developing very fast. The generative adversarial network (GAN) emerges as a promising framework, which uses adversarial training to improve the ...generative ability of its generator. However, since GAN and most of its variants use randomly sampled noises as the input of their generators, they have to learn a mapping function from a whole random distribution to the image manifold. As the structures of the random distribution and the image manifold are generally different, this results in GAN and its variants difficult to train and converge. In this paper, we propose a novel deep model called generative adversarial networks with decoder–encoder output noises (DE-GANs), which take advantage of both the adversarial training and the variational Bayesian inference to improve GAN and its variants on image generation performances. DE-GANs use a pre-trained decoder–encoder architecture to map the random noise vectors to informative ones and feed them to the generator of the adversarial networks. Since the decoder–encoder architecture is trained with the same data set as the generator, its output vectors, as the inputs of the generator, could carry the intrinsic distribution information of the training images, which greatly improves the learnability of the generator and the quality of the generated images. Extensive experiments demonstrate the effectiveness of the proposed model, DE-GANs.
•We propose a new model called GANs with decoder-encoder output noises (DE-GANs).•The decoder-encoder structure learns informative noises as the inputs of DE-GANs.•The informative noises carry the intrinsic information of the image manifold.•DE-GANs converge fast and can generate high quality images.
Class imbalanced datasets are common in real-world applications ranging from credit card fraud detection to rare disease diagnosis. Recently, deep generative models have proved successful for an ...array of machine learning problems such as semi-supervised learning, transfer learning, and recommender systems. However their application to class imbalance situations is limited. In this paper, we consider class conditional variants of generative adversarial networks and variational autoencoders and apply them to the imbalance problem. The main question we seek to answer is whether or not deep conditional generative models can effectively learn the distributions of minority classes so as to produce synthetic observations that ultimately lead to improvements in the performance of a downstream classifier. The numerical results show that this is indeed true and that deep generative models outperform traditional oversampling methods in many circumstances, especially in cases of severe imbalance.
•Propose the conditional Variational Autoencoder for the imbalance problem.•Deep conditional generative models can improve accuracy in imbalanced learning.•Variational Autoencoders achieve better results than Generative Adversarial Networks.•Random oversampling prior to training generative models often lead to better results.
A novel integrated deep learning approach for data-driven surrogate modelling of combustion computational fluid dynamics (CFD) simulations is presented. It combines variational autoencoders (VAEs) ...with deep neural networks (DNNs) to predict detail cell-by-cell two-dimensional distributions of temperature, velocity and species mass fractions from high level inputs such as velocity and fuel and air mass fractions. The VAE model is used to generate low dimensional encodings of the CFD data and the DNN is used in turn to map boundary conditions to the encodings. The results show that regularization is required during all training phases. Sufficiently accurate results were achieved for the reproduced species mass fractions with mean average errors below 0.3 %wt.. The validation mean average percentage errors for the temperature and velocity fields are 1.7% and 7.1% respectively. It is therefore possible to predict detail two-dimensional contours of CFD solution data with adequate generalizability and accuracy.
•Development of a variational-autoencoder capable of compressing CFD data.•Training of deep neural network, which maps boundary conditions to compressed space.•Multiple hyperparameter configurations tested and best-performing model found.•Mean absolute error predictions of species concentrations below 0.3%.•Mean absolute temperature and velocity field prediction errors below 7%.
Microstructure is key to controlling and understanding the properties of materials, but traditional approaches to describing microstructure capture only a small number of features. We require more ...complete descriptors of microstructure to enable data-centric approaches to materials discovery, to allow efficient storage of microstructural data and to assist in quality control in metals processing. The concept of microstructural fingerprinting, using machine learning (ML) to develop quantitative, low-dimensional descriptors of microstructures, has recently attracted significant attention. However, it is difficult to interpret conclusions drawn by ML algorithms, which are often referred to as “black boxes”. For example, convolutional neural networks (CNNs) can be trained to make predictions about a material from a set of microstructural image data, but the feature space that is learned is often used uncritically and adopted without any validation.
Here we explore the use of variational autoencoders (VAEs), comprising a pair of CNNs, which can be trained to produce microstructural fingerprints in a continuous latent space. The VAE architecture also permits the reconstruction of images from fingerprints, allowing us to explore how key features of microstructure are encoded in the latent space of fingerprints. We develop a VAE architecture based on ResNet18 and train it on two classes of Ti-6Al-4V optical micrographs (bimodal and lamellar) as an example of an industrially important alloy where microstructural control is critical to performance. The latent/feature space of fingerprints learned by the VAE is explored in several ways, including by supplying interpolated and randomly perturbed fingerprints to the trained decoder and via dimensionality reduction to explore the distribution and correlation of microstructural features within the latent space of fingerprints.
We demonstrate that the fingerprints generated via the trained VAE exhibit smooth, interpolable behaviour with stability to local perturbations, supporting their suitability as general purpose descriptors for microstructure. The analysis of computational results uncover that key properties of the microstructures (volume fraction and grain size) are strongly correlated with position in the encoded feature space, supporting the use of VAE fingerprints for quantitative exploration of process–structure–property relationships.
Display omitted
•VAE based on ResNet is able to produce accurate reconstructions of microstructures.•PCA provides useful insight into the boundaries of validity in the latent space.•t-SNE applied to latent space to visualise morphological feature distributions.•VAE fingerprints shown to correlate with morphological features via regression.
Lifelong Mixture of Variational Autoencoders Ye, Fei; Bors, Adrian G.
IEEE transaction on neural networks and learning systems,
2023-Jan., 2023-Jan, 2023-1-00, 20230101, Volume:
34, Issue:
1
Journal Article
Open access
In this article, we propose an end-to-end lifelong learning mixture of experts. Each expert is implemented by a variational autoencoder (VAE). The experts in the mixture system are jointly trained by ...maximizing a mixture of individual component evidence lower bounds (MELBO) on the log-likelihood of the given training samples. The mixing coefficients in the mixture model control the contributions of each expert in the global representation. These are sampled from a Dirichlet distribution whose parameters are determined through nonparametric estimation during lifelong learning. The model can learn new tasks fast when these are similar to those previously learned. The proposed lifelong mixture of VAE (L-MVAE) expands its architecture with new components when learning a completely new task. After the training, our model can automatically determine the relevant expert to be used when fed with new data samples. This mechanism benefits both the memory efficiency and the required computational cost as only one expert is used during the inference. The L-MVAE inference model is able to perform interpolations in the joint latent space across the data domains associated with different tasks and is shown to be efficient for disentangled learning representation.
Graph autoencoders (AE) and variational autoencoders (VAE) are powerful node embedding methods, but suffer from scalability issues. In this paper, we introduce FastGAE, a general framework to scale ...graph AE and VAE to large graphs with millions of nodes and edges. Our strategy, based on an effective stochastic subgraph decoding scheme, significantly speeds up the training of graph AE and VAE while preserving or even improving performances. We demonstrate the effectiveness of FastGAE on various real-world graphs, outperforming the few existing approaches to scale graph AE and VAE by a wide margin.
The inherent ambiguity in ground-truth annotations of 3D bounding boxes, caused by occlusions, signal missing, or manual annotation errors, can confuse deep 3D object detectors during training, thus ...deteriorating detection accuracy. However, existing methods overlook such issues to some extent and treat the labels ass deterministic. In this paper, we formulate the label uncertainty problem as the diversity of potentially plausible bounding boxes of objects. Then, we propose GLENet, a generative framework adapted from conditional variational autoencoders, to model the one-to-many relationship between a typical 3D object and its potential ground-truth bounding boxes with latent variables. The label uncertainty generated by GLENet is a plug-and-play module and can be conveniently integrated into existing deep 3D detectors to build probabilistic detectors and supervise the learning of the localization uncertainty. Besides, we propose an uncertainty-aware quality estimator architecture in probabilistic detectors to guide the training of the IoU-branch with predicted localization uncertainty. We incorporate the proposed methods into various popular base 3D detectors and demonstrate significant and consistent performance gains on both KITTI and Waymo benchmark datasets. Especially, the proposed GLENet-VR outperforms all published LiDAR-based approaches by a large margin and achieves the top rank among single-modal methods on the challenging KITTI test set. The source code and pre-trained models are publicly available at
https://github.com/Eaphan/GLENet
.
Deep Mixture Generative Autoencoders Ye, Fei; Bors, Adrian G.
IEEE transaction on neural networks and learning systems,
10/2022, Volume:
33, Issue:
10
Journal Article
Open access
Variational autoencoders (VAEs) are one of the most popular unsupervised generative models that rely on learning latent representations of data. In this article, we extend the classical concept of ...Gaussian mixtures into the deep variational framework by proposing a mixture of VAEs (MVAE). Each component in the MVAE model is implemented by a variational encoder and has an associated subdecoder. The separation between the latent spaces modeled by different encoders is enforced using the <inline-formula> <tex-math notation="LaTeX">d </tex-math></inline-formula>-variable Hilbert-Schmidt independence criterion (dHSIC). Each component would capture different data variational features. We also propose a mechanism for finding the appropriate number of VAE components for a given task, leading to an optimal architecture. The differentiable categorical Gumbel-softmax distribution is used in order to generate dropout masking parameters within the end-to-end backpropagation training framework. Extensive experiments show that the proposed MVAE model can learn a rich latent data representation and is able to discover additional underlying data representation factors.