Risk-Averse Explore-Then-Commit Algorithms for Finite-Time Bandits

In this paper, we study multi-armed bandit problems in explore-then-commit setting. In our proposed explore-then-commit setting, the goal is to identify the best arm after a pure experimentation (exploration) phase and exploit it once or for a given finite number of times. We identify that although the arm with the highest expected reward is the most desirable objective for infinite exploitations, it is not necessarily the one that is most probable to have the highest reward in a single or finite-time exploitations. Alternatively, we advocate the idea of risk-aversion where the objective is to compete against the arm with the best risk-return trade-off. Then, we propose two algorithms whose objectives are to select the arm that is most probable to reward the most. Using a new notion of finite-time exploitation regret, we find an upper bound for the minimum number of experiments before commitment, to guarantee an upper bound for the regret. As compared to existing risk-averse bandit algorithms, our algorithms do not rely on hyper-parameters, resulting in a more robust behavior in practice, which is verified by the numerical evaluation.

MONCAE: Multi-Objective Neuroevolution of Convolutional Autoencoders

In this paper, we present a novel neuroevolutionary method to identify the architecture and hyperparameters of convolutional autoencoders. Remarkably, we used a hypervolume indicator in the context of neural architecture search for autoencoders, for the first time to our current knowledge. Results show that images were compressed by a factor of more than 10, while still retaining enough information to achieve image classification for the majority of the tasks. Thus, this new approach can be used to speed up the AutoML pipeline for image compression.

Exact Acceleration of K-Means++ and K-Means$\|$

K-Means++ and its distributed variant K-Means$\|$ have become de facto tools for selecting the initial seeds of K-means. While alternatives have been developed, the effectiveness, ease of implementation, and theoretical grounding of the K-means++ and $\|$ methods have made them difficult to "best" from a holistic perspective. By considering the limited opportunities within seed selection to perform pruning, we develop specialized triangle inequality pruning strategies and a dynamic priority queue to show the first acceleration of K-Means++ and K-Means$\|$ that is faster in run-time while being algorithmicly equivalent. For both algorithms we are able to reduce distance computations by over $500\times$. For K-means++ this results in up to a 17$\times$ speedup in run-time and a $551\times$ speedup for K-means$\|$. We achieve this with simple, but carefully chosen, modifications to known techniques which makes it easy to integrate our approach into existing implementations of these algorithms.

Long short-term relevance learning

To incorporate prior knowledge as well as measurement uncertainties in the traditional long short term memory (LSTM) neural networks, an efficient sparse Bayesian training algorithm is introduced to the network architecture. The proposed scheme automatically determines relevant neural connections and adapts accordingly, in contrast to the classical LSTM solution. Due to its flexibility, the new LSTM scheme is less prone to overfitting, and hence can approximate time dependent solutions by use of a smaller data set. On a structural nonlinear finite element application we show that the self-regulating framework does not require prior knowledge of a suitable network architecture and size, while ensuring satisfying accuracy at reasonable computational cost.

Model-Parallel Model Selection for Deep Learning Systems

As deep learning becomes more expensive, both in terms of time and compute, inefficiencies in machine learning (ML) training prevent practical usage of state-of-the-art models for most users. The newest model architectures are simply too large to be fit onto a single processor. To address the issue, many ML practitioners have turned to model parallelism as a method of distributing the computational requirements across several devices. Unfortunately, the sequential nature of neural networks causes very low efficiency and device utilization in model parallel training jobs. We propose a new form of "shard parallelism" combining task and model parallelism, then package it into a framework we name Hydra. Hydra recasts the problem of model parallelism in the multi-model context to produce a fine-grained parallel workload of independent model shards, rather than independent models. This new parallel design promises dramatic speedups relative to the traditional model parallelism paradigm.

Modelling the Time-dependent VRP through Open Data

This paper presents an open data approach to model and solve the vehicle routing problem with time-dependent travel times (TDVRP). The proposed model is based on a multi-layer matrix composed of travel times, replacing the traditional distance matrix. Online cartography services are queried in order to build this matrix. Travel times are obtained for every step in the time discretization. Thus, the model integrates the fact that the travel time between two points is modified during the time horizon. This model is applied to a medium-sized problem in the urban area of Paris using an enhanced Greedy Randomized Adaptive Search Procedure (GRASP). This work intends to build on the current state of the art by proposing a straightforward and open-access method to model and solve the VRP with traffic variability.

Optimized multi-axis spiral projection MR fingerprinting with subspace reconstruction for rapid whole-brain high-isotropic-resolution quantitative imaging

Purpose: To improve image quality and accelerate the acquisition of 3D MRF. Methods: Building on the multi-axis spiral-projection MRF technique, a subspace reconstruction with locally low rank (LLR) constraint and a modified spiral-projection spatiotemporal encoding scheme termed tiny-golden-angle-shuffling (TGAS) were implemented for rapid whole-brain high-resolution quantitative mapping. The LLR regularization parameter and the number of subspace bases were tuned using retrospective in-vivo data and simulated examinations, respectively. B0 inhomogeneity correction using multi-frequency interpolation was incorporated into the subspace reconstruction to further improve the image quality by mitigating blurring caused by off-resonance effect. Results: The proposed MRF acquisition and reconstruction framework can produce provide high quality 1-mm isotropic whole-brain quantitative maps in a total acquisition time of 1 minute 55 seconds, with higher-quality results than ones obtained from the previous approach in 6 minutes. The comparison of quantitative results indicates that neither the subspace reconstruction nor the TGAS trajectory induce bias for T1 and T2 mapping. High quality whole-brain MRF data were also obtained at 0.66-mm isotropic resolution in 4 minutes using the proposed technique, where the increased resolution was shown to improve visualization of subtle brain structures. Conclusion: The proposed TGAS-SPI-MRF with optimized spiral-projection trajectory and subspace reconstruction can enable high-resolution quantitative mapping with faster acquisition speed.

Agile wide-field imaging with selective high resolution

Wide-field and high-resolution (HR) imaging is essential for various applications such as aviation reconnaissance, topographic mapping and safety monitoring. The existing techniques require a large-scale detector array to capture HR images of the whole field, resulting in high complexity and heavy cost. In this work, we report an agile wide-field imaging framework with selective high resolution that requires only two detectors. It builds on the statistical sparsity prior of natural scenes that the important targets locate only at small regions of interests (ROI), instead of the whole field. Under this assumption, we use a short-focal camera to image wide field with a certain low resolution, and use a long-focal camera to acquire the HR images of ROI. To automatically locate ROI in the wide field in real time, we propose an efficient deep-learning based multiscale registration method that is robust and blind to the large setting differences (focal, white balance, etc) between the two cameras. Using the registered location, the long-focal camera mounted on a gimbal enables real-time tracking of the ROI for continuous HR imaging. We demonstrated the novel imaging framework by building a proof-of-concept setup with only 1181 gram weight, and assembled it on an unmanned aerial vehicle for air-to-ground monitoring. Experiments show that the setup maintains 120$^{\circ}$ wide field-of-view (FOV) with selective 0.45$mrad$ instantaneous FOV.

FastS2S-VC: Streaming Non-Autoregressive Sequence-to-Sequence Voice Conversion

This paper proposes a non-autoregressive extension of our previously proposed sequence-to-sequence (S2S) model-based voice conversion (VC) methods. S2S model-based VC methods have attracted particular attention in recent years for their flexibility in converting not only the voice identity but also the pitch contour and local duration of input speech, thanks to the ability of the encoder-decoder architecture with the attention mechanism. However, one of the obstacles to making these methods work in real-time is the autoregressive (AR) structure. To overcome this obstacle, we develop a method to obtain a model that is free from an AR structure and behaves similarly to the original S2S models, based on a teacher-student learning framework. In our method, called "FastS2S-VC", the student model consists of encoder, decoder, and attention predictor. The attention predictor learns to predict attention distributions solely from source speech along with a target class index with the guidance of those predicted by the teacher model from both source and target speech. Thanks to this structure, the model is freed from an AR structure and allows for parallelization. Furthermore, we show that FastS2S-VC is suitable for real-time implementation based on a sliding-window approach, and describe how to make it run in real-time. Through speaker-identity and emotional-expression conversion experiments, we confirmed that FastS2S-VC was able to speed up the conversion process by 70 to 100 times compared to the original AR-type S2S-VC methods, without significantly degrading the audio quality and similarity to target speech. We also confirmed that the real-time version of FastS2S-VC can be run with a latency of 32 ms when run on a GPU.

Automatic Brain Tumor Segmentation using Convolutional Neural Networks with Test-Time Augmentation

Automatic brain tumor segmentation plays an important role for diagnosis, surgical planning and treatment assessment of brain tumors. Deep convolutional neural networks (CNNs) have been widely used for this task. Due to the relatively small data set for training, data augmentation at training time has been commonly used for better performance of CNNs. Recent works also demonstrated the usefulness of using augmentation at test time, in addition to training time, for achieving more robust predictions. We investigate how test-time augmentation can improve CNNs' performance for brain tumor segmentation. We used different underpinning network structures and augmented the image by 3D rotation, flipping, scaling and adding random noise at both training and test time. Experiments with BraTS 2018 training and validation set show that test-time augmentation helps to improve the brain tumor segmentation accuracy and obtain uncertainty estimation of the segmentation results.