The success of deep (reinforcement) learning systems crucially depends on the correct choice of hyperparameters which are notoriously sensitive and expensive to evaluate. Training these systems typically requires running iterative processes over multiple epochs or episodes. Traditional approaches only consider final performances of a hyperparameter although intermediate information from the learning curve is readily available. In this paper, we present a Bayesian optimization approach which exploits the iterative structure of learning algorithms for efficient hyperparameter tuning. First, we transform each training curve into a numeric score. Second, we selectively augment the data using the auxiliary information from the curve. This augmentation step enables modeling efficiency while preventing the ill-conditioned issue of Gaussian process covariance matrix happened when adding the whole curve. We demonstrate the efficiency of our algorithm by tuning hyperparameters for the training of deep reinforcement learning agents and convolutional neural networks. Our algorithm outperforms all existing baselines in identifying optimal hyperparameters in minimal time.

Practical reinforcement learning problems are often formulated as constrained Markov decision process (CMDP) problems, in which the agent has to maximize the expected return while satisfying a set of prescribed safety constraints. In this study, we propose a novel simulator-based method to approximately solve a CMDP problem without making any compromise on the safety constraints. We achieve this by decomposing the CMDP into a pair of MDPs; reconnaissance MDP and planning MDP. The purpose of reconnaissance MDP is to evaluate the set of actions that are safe, and the purpose of planning MDP is to maximize the return while using the actions authorized by reconnaissance MDP. RMDP can define a set of safe policies for any given set of safety constraint, and this set of safe policies can be used to solve another CMDP problem with different reward. Our method is not only computationally less demanding than the previous simulator-based approaches to CMDP, but also capable of finding a competitive reward-seeking policy in a high dimensional environment, including those involving multiple moving obstacles.

We introduce a framework to study the transformation of problems with manifold constraints into unconstrained problems through parametrizations in terms of a Euclidean space. We call these parametrizations "trivializations". We prove conditions under which a trivialization is sound in the context of gradient-based optimization and we show how two large families of trivializations have overall favorable properties, but also suffer from a performance issue. We then introduce "dynamic trivializations", which solve this problem, and we show how these form a family of optimization methods that lie between trivializations and Riemannian gradient descent, and combine the benefits of both of them. We then show how to implement these two families of trivializations in practice for different matrix manifolds. To this end, we prove a formula for the gradient of the exponential of matrices, which can be of practical interest on its own. Finally, we show how dynamic trivializations improve the performance of existing methods on standard tasks designed to test long-term memory within neural networks.

Furthermore it has provided theoretical certitude on the effectivness of the procedure. However, the iterative optimisation method associated with the diversity and flexibility of the architecture of a firing graph opens doors to further applications, notably in the field of inverse problem and in the very hype field of machine learning. Indeed in supervised classification, we are given a dataset composed of features that may be numerical or categorical description of samples and targets that specify the class of samples. If we assume that the activation of a target is a combination of latent factors's activations and that we operate the minimum transformation of features so that they take the form of a measure grid, a light layer of procedures could turn our solution into a supervised classificator. The specificity of such a learner would give it an interesting position in the supervised learning landscape. Indeed, its iterative optimisation and flexible architecture could make it an adaptative learner, that scale to large dataset, with minimum processing work on raw data, in the manner of a neural network. Yet unlike neural network the algorithm handle very efficiently categorical or sparse feature space. Furthermore, compared to the most advanced tree based classification, its flexible architecture is more suitable to learning update and on-the-fly evaluation or addition of new features. Finally, given the hype granted to the field of machine learning nowaday, both in the scientific comunity and civil society, it would be common sense to orient this piece of research to this field.

We propose a multi-level method to increase the accuracy of machine learning algorithms for approximating observables in scientific computing, particularly those that arise in systems modeled by differential equations. The algorithm relies on judiciously combining a large number of computationally cheap training data on coarse resolutions with a few expensive training samples on fine grid resolutions. Theoretical arguments for lowering the generalization error, based on reducing the variance of the underlying maps, are provided and numerical evidence, indicating significant gains over underlying single-level machine learning algorithms, are presented. Moreover, we also apply the multi-level algorithm in the context of forward uncertainty quantification and observe a considerable speed-up over competing algorithms.

Metaheuristic algorithms (MAs) have seen unprecedented growth thanks to their successful applications in fields including engineering and health sciences. In this work, we investigate the use of a deep learning (DL) model as an alternative tool to do so. The proposed method, called MaNet, is motivated by the fact that most of the DL models often need to solve massive nasty optimization problems consisting of millions of parameters. Feature selection is the main adopted concepts in MaNet that helps the algorithm to skip irrelevant or partially relevant evolutionary information and uses those which contribute most to the overall performance. The introduced model is applied on several unimodal and multimodal continues problems. The experiments indicate that MaNet is able to yield competitive results compared to one of the best hand-designed algorithms for the aforementioned problems, in terms of the solution accuracy and scalability.

Semantic code search is the task of retrieving relevant code given a natural language query. While related to other information retrieval tasks, it requires bridging the gap between the language used in code (often abbreviated and highly technical) and natural language more suitable to describe vague concepts and ideas. To enable evaluation of progress on code search, we are releasing the CodeSearchNet Corpus and are presenting the CodeSearchNet Challenge, which consists of 99 natural language queries with about 4k expert relevance annotations of likely results from CodeSearchNet Corpus. The corpus contains about 6 million functions from open-source code spanning six programming languages (Go, Java, JavaScript, PHP, Python, and Ruby). The CodeSearchNet Corpus also contains automatically generated query-like natural language for 2 million functions, obtained from mechanically scraping and preprocessing associated function documentation. In this article, we describe the methodology used to obtain the corpus and expert labels, as well as a number of simple baseline solutions for the task. We hope that CodeSearchNet Challenge encourages researchers and practitioners to study this interesting task further and will host a competition and leaderboard to track the progress on the challenge. We are also keen on extending \task to more queries and programming languages in the future.

--The purpose of this work is to develop and study a distributed strategy for Pareto optimization of an aggregate cost consisting of regularized risks. Each risk is modeled as the expectation of some loss function with unknown probability distribution while the regularizers are assumed deterministic, but are not required to be differentiable or even continuous. The individual, regularized, cost functions are distributed across a strongly-connected network of agents and the Pareto optimal solution is sought by appealing to a multi-agent diffusion strategy. T o this end, the regularizers are smoothed by means of infimal convolution and it is shown that the Pareto solution of the approximate, smooth problem can be made arbitrarily close to the solution of the original, non-smooth problem. Performance bounds are established under conditions that are weaker than assumed before in the literature, and hence applicable to a broader class of adaptation and learning problems. Index T erms --Distributed optimization, diffusion strategy, smoothing, proximal operator, non-smooth regularizer, proximal diffusion, regularized diffusion. The objective of distributed learning is the solution of global, stochastic optimization problems across networks of agents through localized interactions and without information about the statistical properties of the data. Using streaming data, the resulting strategies are adaptive in nature and able to track drifts in the location of the minimizers due to variations in the statistical properties of the data. Regularization is one useful technique to encourage or enforce structural properties on the sought after minimizer, such as sparsity or constraints. A substantial number of regularizers are inherently non-smooth, while many cost functions are differentiable.

The paper demonstrates the use of the fully convolutional neural network for glioma segmentation on the BraTS 2019 dataset. Three-layers deep encoder-decoder architecture is used along with dense connection at encoder part to propagate the information from coarse layer to deep layers. This architecture is used to train three tumor sub-components separately. Subcomponent training weights are initialized with whole tumor weights to get the localization of the tumor within the brain. At the end, three segmentation results were merged to get the entire tumor segmentation. Dice Similarity of training dataset with focal loss implementation for whole tumor, tumor core and enhancing tumor is 0.92, 0.90 and 0.79 respectively. Radiomic features along with segmentation results and age are used to predict the overall survival of patients using random forest regressor to classify survival of patients in long, medium and short survival classes.

A new procedure to construct predictive models in supervised learning problems by paying attention to the clustering structure of the input data is introduced. We are interested in situations where the input data consists of more than one unknown cluster, and where there exist different underlying models on these clusters. Thus, instead of constructing a single predictive model on the whole dataset, we propose to use a K-means clustering algorithm with different options of Bregman divergences, to recover the clustering structure of the input data. Then one dedicated predictive model is fit per cluster. For each divergence, we construct a simple local predictor on each observed cluster. We obtain one estimator, the collection of the K simple local predictors, per divergence, and we propose to combine them in a smart way based on a consensus idea. Several versions of consensual aggregation in both classification and regression problems are considered. A comparison of the performances of all constructed estimators on different simulated and real data assesses the excellent performance of our method. In a large variety of prediction problems, the consensual aggregation procedure outperforms all the other models.