Compressing deep neural networks (DNNs) is important for real-world applications operating on resource-constrained devices. However, it is difficult to change the model size once the training is completed, which needs retraining to configure models suitable for different devices. In this paper, we propose a novel method that enables DNNs to flexibly change their size after training. We factorize the weight matrices of the DNNs via singular value decomposition (SVD) and change their ranks according to the target size. In contrast with existing methods, we introduce simple criteria that characterize the importance of each basis and layer, which enables to effectively compress the error and complexity of models as little as possible. In experiments on multiple image-classification tasks, our method exhibits favorable performance compared with other methods. As part of the great progress made in deep learning, deep neural network (DNN) models with higher performance have been proposed for various machine-learning tasks (LeCun et al., 2015). However, these performance improvements require a higher number of parameters and greater computational complexity. Therefore, it is important to compress them without sacrificing the performance for running the models on resource-constrained devices. Han et al. (2016) reduced the memory requirement for devices by pruning and quantizing weight coefficients after training the models.

--Evaluating, explaining, and visualizing high-level concepts in generative models, such as variational autoencoders (V AEs), is challenging in part due to a lack of known prediction classes that are required to generate saliency maps in supervised learning. While saliency maps may help identify relevant features (e.g., pixels) in the input for classification tasks of deep neural networks, similar frameworks are understudied in unsupervised learning. Therefore, we introduce a new method of obtaining saliency maps for latent representations of known or novel high-level concepts, often called concept vectors in generative models. Concept scores, analogous to class scores in classification tasks, are defined as dot products between concept vectors and encoded input data, which can be readily used to compute the gradients. The resulting concept saliency maps are shown to highlight input features deemed important for high-level concepts. Our method is applied to the V AE's latent space of CelebA dataset in which known attributes such as "smiles" and "hats" are used to elucidate relevant facial features. Furthermore, our application to spatial transcriptomic (ST) data of a mouse olfactory bulb demonstrates the potential of latent representations of morphological layers and molecular features in advancing our understanding of complex biological systems. By extending the popular method of saliency maps to generative models, the proposed concept saliency maps help improve interpretability of latent variable models in deep learning. I NTRODUCTION A rapidly increasing amount of unlabeled data, such as images and molecular data, has prompted a rise of deep generative models, that can be trained without human supervision. By using a vast amount of unlabeled data, unsupervised learning models such as variational autoencoders (V AEs) [1], [2] extract low-dimensional latent spaces that compactly encode high-dimensional input data and potentially reveal hidden relationships.

Prediction of molecular properties, including physico-chemical properties, is a challenging task in chemistry. Herein we present a new state-of-the-art multitask prediction method based on existing graph neural network methods. We have used different architectures for our models and the results clearly demonstrate that multitask learning can improve model performance. Additionally, a significant reduction of variance in the models have been observed. Most importantly, datasets with a small amount of data points reach better results without the need of augmentation. The improvement is an effect of data augmentation.

This paper tests the hypothesis that modeling a scene in terms of entities and their local interactions, as opposed to modeling the scene globally, provides a significant benefit in generalizing to physical tasks in a combinatorial space the learner has not encountered before. We present object-centric perception, prediction, and planning (OP3), which to the best of our knowledge is the first entity-centric dynamic latent variable framework for model-based reinforcement learning that acquires entity representations from raw visual observations without supervision and uses them to predict and plan. OP3 enforces entity-abstraction -- symmetric processing of each entity representation with the same locally-scoped function -- which enables it to scale to model different numbers and configurations of objects from those in training. Our approach to solving the key technical challenge of grounding these entity representations to actual objects in the environment is to frame this variable binding problem as an inference problem, and we developing an interactive inference algorithm that uses temporal continuity and interactive feedback to bind information about object properties to the entity variables. On block-stacking tasks, OP3 generalizes to novel block configurations and more objects than observed during training, outperforming an oracle model that assumes access to object supervision and achieving two to three times better accuracy than a state-of-the-art video prediction model.

David Obst a,b, Badih Ghattas b, Sandra Claudel a, Jairo Cugliari c, Yannig Goude a, Georges Oppenheim d a EDF R&D, Palaiseau, France b Institut de Math ematiques de Marseille, Aix-Marseille Universit e, France c ERIC, Universit e de Lyon 2, France d Laboratoire d'Analyse et de Math ematiques Appliqu ees Universit e Paris-Est, Champs-sur-Marne, FranceAbstract While ubiquitous, textual sources of information such as company reports, social media posts, etc. are hardly included in prediction algorithms for time series, despite the relevant information they may contain. In this work, openly accessible daily weather reports from France and the United-Kingdom are leveraged to predict time series of national electricity consumption, average temperature and wind-speed with a single pipeline. Two methods of numerical representation of text are considered, namely traditional Term Frequency - Inverse Document Frequency (TF-IDF) as well as our own neural word embedding. Using exclusively text, we are able to predict the aforementioned time series with sufficient accuracy to be used to replace missing data. Furthermore the proposed word embeddings display geometric properties relating to the behavior of the time series and context similarity between words. Introduction Whether it is in the field of energy, finance or meteorology, accurately predicting the behavior of time series is nowadays of paramount importance for optimal decision making or profit. While the field of time series forecasting is extremely prolific from a research point-of-view, up to now it has narrowed its efforts on the exploitation of regular numerical features extracted from sensors, data bases or stock exchanges. Unstructured data such as text on the other hand remains underexploited for prediction tasks, despite its potentially valuable informative content. Empirical studies have already proven that textual sources such as news articles or blog entries can be correlated to stock exchange time series and have explanatory power for their variations [1, 2]. This observation has motivated multiple extensive experiments to extract relevant features from textual documents in different ways and use them for prediction, notably in the field of finance. In Lavrenko et al. [3], language models (considering only the presence of a word) are used to estimate the probability of trends such as surges or falls of 127 different stock values using articles from Biz Yahoo!. Their results show that this text driven approach could be used to make profit on the market. One of the most conventional ways for text representation is the TF-IDF (Term Frequency - Inverse Document Frequency) approach.

In regression tasks, aleatoric uncertainty is commonly addressed by considering a parametric distribution of the output variable, which is based on strong assumptions such as symmetry, unimodality or by supposing a restricted shape. These assumptions are too limited in scenarios where complex shapes, strong skews or multiple modes are present. In this paper, we propose a generic deep learning framework that learns an Uncountable Mixture of Asymmetric Laplacians (UMAL), which will allow us to estimate heterogeneous distributions of the output variable and shows its connections to quantile regression. Despite having a fixed number of parameters, the model can be interpreted as an infinite mixture of components, which yields a flexible approximation for heterogeneous distributions. Apart from synthetic cases, we apply this model to room price forecasting and to predict financial operations in personal bank accounts. We demonstrate that UMAL produces proper distributions, which allows us to extract richer insights and to sharpen decision-making.

Artificial neural network has achieved unprecedented success in the medical domain. This success depends on the availability of massive and representative datasets. However, data collection is often prevented by privacy concerns and people want to take control over their sensitive information during both training and using processes. To address this problem, we propose a privacy-preserving method for the distributed system, Stochastic Channel-Based Federated Learning (SCBF), which enables the participants to train a high-performance model cooperatively without sharing their inputs. Specifically, we design, implement and evaluate a channel-based update algorithm for the central server in a distributed system, which selects the channels with regard to the most active features in a training loop and uploads them as learned information from local datasets. A pruning process is applied to the algorithm based on the validation set, which serves as a model accelerator. In the experiment, our model presents better performances and higher saturating speed than the Federated Averaging method which reveals all the parameters of local models to the server when updating. We also demonstrate that the saturating rate of performance could be promoted by introducing a pruning process. And further improvement could be achieved by tuning the pruning rate. Our experiment shows that 57% of the time is saved by the pruning process with only a reduction of 0.0047 in AUCROC performance and a reduction of 0.0068 in AUCPR.

This paper investigates statistical models for road traffic modeling. The proposed methodology considers road traffic as a (i) highdimensional time-series for which (ii) regeneration occurs at the end of each day. Since (ii), prediction is based on a daily modeling of the road traffic using a vector autoregressive model that combines linearly the past observations of the day. Considering (i), the learning algorithm follows from an l1-penalization of the regression coefficients. Excess risk bounds are established under the high-dimensional framework in which the number of road sections goes to infinity with the number of observed days. Considering floating car data observed in an urban area, the approach is compared to state-of-the-art methods including neural networks. In addition of being very competitive in terms of prediction, it enables to identify the most determinant sections of the road network.

Stochastic gradient descent (SGD) forms the core optimization method for deep neural networks. While some theoretical progress has been made, it still remains unclear why SGD leads the learning dynamics in overparameterized networks to solutions that generalize well. Here we show that for overparameterized networks with a degenerate valley in their loss landscape, SGD on average decreases the trace of the Hessian of the loss. We also generalize this result to other noise structures and show that isotropic noise in the non-degenerate subspace of the Hessian decreases its determinant. In addition to explaining SGDs role in sculpting the Hessian spectrum, this opens the door to new optimization approaches that may confer better generalization performance. We test our results with experiments on toy models and deep neural networks.

Here comes the best part: training and getting our actual results! First off, we'll need to do a bit of data preprocessing to have the data properly formatted for training.