"No free lunch" results state the impossibility of obtaining meaningful bounds on the error of a learning algorithm without prior assumptions and modelling. Some models are expensive (strong assumptions, such as as subgaussian tails), others are cheap (simply finite variance). As it is well known, the more you pay, the more you get: in other words, the most expensive models yield the more interesting bounds. Recent advances in robust statistics have investigated procedures to obtain tight bounds while keeping the cost minimal. The present paper explores and exhibits what the limits are for obtaining tight PAC-Bayes bounds in a robust setting for cheap models, addressing the question: is PAC-Bayes good value for money?

A common difficulty in applications of machine learning is the lack of any general principle for guiding the choices of key parameters of the underlying neural network. Focusing on a class of recurrent neural networks - reservoir computing systems that have recently been exploited for model-free prediction of nonlinear dynamical systems, we uncover a surprising phenomenon: the emergence of an interval in the spectral radius of the neural network in which the prediction error is minimized. In a three-dimensional representation of the error versus time and spectral radius, the interval corresponds to the bottom region of a "valley." Such a valley arises for a variety of spatiotemporal dynamical systems described by nonlinear partial differential equations, regardless of the structure and the edge-weight distribution of the underlying reservoir network. We also find that, while the particular location and size of the valley would depend on the details of the target system to be predicted, the interval tends to be larger for undirected than for directed networks. The valley phenomenon can be beneficial to the design of optimal reservoir computing, representing a small step forward in understanding these machine-learning systems.

In real world machine learning applications, testing data may contain some meaningful new categories that have not been seen in labeled training data. To simultaneously recognize new data categories and assign most appropriate category labels to the data actually from known categories, existing models assume the number of unknown new categories is pre-specified, though it is difficult to determine in advance. In this paper, we propose a Bayesian nonparametric topic model to automatically infer this number, based on the hierarchical Dirichlet process and the notion of latent Dirichlet allocation. Exact inference in our model is intractable, so we provide an efficient collapsed Gibbs sampling algorithm for approximate posterior inference. Extensive experiments on various text data sets show that: (a) compared with parametric approaches that use pre-specified true number of new categories, the proposed nonparametric approach can yield comparable performance; and (b) when the exact number of new categories is unavailable, i.e. the parametric approaches only have a rough idea about the new categories, our approach has evident performance advantages.

DOA ESTIMA TION BY DNN-BASED DENOISING AND DEREVERBERA TION FROM SOUND INTENSITY VECTOR Masahiro Y asuda 1, Y uma Koizumi 1, Luca Mazzon 2, Shoichiro Saito 1 and Hisashi Uematsu 1 1 NTT Media Intelligence Laboratories, Tokyo, Japan 2 University of Padova, Padua, Italy ABSTRACT We propose a direction of arrival (DOA) estimation method that combines sound-intensity vector (IV)-based DOA estimation and DNN-based denoising and dereverberation. Since the accuracy of IV -based DOA estimation degrades due to environmental noise and reverberation, two DNNs are used to remove such effects from the observed IVs. DOA is then estimated from the refined IVs based on the physics of wave propagation. Experiments on an open dataset showed that the average DOA error of the proposed method was 0.528 degrees, and it outperformed a conventional IV -based and DNN-based DOA estimation method. Index T erms-- direction of arrival, deep neural network, sound intensity vector, sound activity detection 1. INTRODUCTION Time series direction-of-arrival (DOA) estimation, which is the task of identifying the relative position of the sound sources with respect to the microphone at every time frame, is an important technology for understanding the surrounding environment from sound recordings. For example, DOA estimation is useful for autonomous driving that autonomously acquiring the surrounding environment [1].

Weakly-supervised learning is a paradigm for alleviating the scarcity of labeled data by leveraging lower-quality but larger-scale supervision signals. While existing work mainly focuses on utilizing a certain type of weak supervision, we present a probabilistic framework, learning from indirect observations, for learning from a wide range of weak supervision in real-world problems, e.g., noisy labels, complementary labels and coarse-grained labels. We propose a general method based on the maximum likelihood principle, which has desirable theoretical properties and can be straightforwardly implemented for deep neural networks. Concretely, a discriminative model for the true target is used for modeling the indirect observation, which is a random variable entirely depending on the true target stochastically or deterministically. Then, maximizing the likelihood given indirect observations leads to an estimator of the true target implicitly. Comprehensive experiments for two novel problem settings --- learning from multiclass label proportions and learning from coarse-grained labels, illustrate practical usefulness of our method and demonstrate how to integrate various sources of weak supervision.

In this paper, we propose a novel data augmentation method for training neural networks for Direction of Arrival (DOA) estimation. This method focuses on expanding the representation of the DOA subspace of a dataset. Given some input data, it applies a transformation to it in order to change its DOA information and simulate new potentially unseen one. Such transformation, in general, is a combination of a rotation and a reflection. It is possible to apply such transformation due to a well-known property of First Order Ambisonics (FOA). The same transformation is applied also to the labels, in order to maintain consistency between input data and target labels. Three methods with different level of generality are proposed for applying this augmentation principle. Experiments are conducted on two different DOA networks. Results of both experiments demonstrate the effectiveness of the novel augmentation strategy by improving the DOA error by around 40%.

We consider a document classification problem where document labels are absent but only relevant keywords of a target class and unlabeled documents are given. Although heuristic methods based on pseudo-labeling have been considered, theoretical understanding of this problem has still been limited. Moreover, previous methods cannot easily incorporate well-developed techniques in supervised text classification. In this paper, we propose a theoretically guaranteed learning framework that is simple to implement and has flexible choices of models, e.g., linear models or neural networks. We demonstrate how to optimize the area under the receiver operating characteristic curve (AUC) effectively and also discuss how to adjust it to optimize other well-known evaluation metrics such as the accuracy and F1-measure. Finally, we show the effectiveness of our framework using benchmark datasets.

This paper considers a variant of the classical online learning problem with expert predictions. Our model's differences and challenges are due to lacking any direct feedback on the loss each expert incurs at each time step $t$. We propose an approach that uses peer assessment and identify conditions where it succeeds. Our techniques revolve around a carefully designed peer score function $s()$ that scores experts' predictions based on the peer consensus. We show a sufficient condition, that we call \emph{peer calibration}, under which standard online learning algorithms using loss feedback computed by the carefully crafted $s()$ have bounded regret with respect to the unrevealed ground truth values. We then demonstrate how suitable $s()$ functions can be derived for different assumptions and models.

We consider an active learning setting where the algorithm has access to a large pool of unlabeled data and a small pool of labeled data. In each iteration, the algorithm chooses few unlabeled data points and obtains their labels from an oracle. In this paper, we consider a probabilistic querying procedure to choose the points to be labeled. We propose an algorithm for Active Learning with Importance Sampling (ALIS), and derive upper bounds on the true loss incurred by the algorithm for any arbitrary probabilistic sampling procedure. Further, we propose an optimal sampling distribution that minimizes the upper bound on the true loss.

Optimal transport distances are powerful tools to compare probability distributions and have found many applications in machine learning. Yet their algorithmic complexity prevents their direct use on large scale datasets. To overcome this challenge, practitioners compute these distances on minibatches {\em i.e.} they average the outcome of several smaller optimal transport problems. We propose in this paper an analysis of this practice, which effects are not well understood so far. We notably argue that it is equivalent to an implicit regularization of the original problem, with appealing properties such as unbiased estimators, gradients and a concentration bound around the expectation, but also with defects such as loss of distance property. Along with this theoretical analysis, we also conduct empirical experiments on gradient flows, GANs or color transfer that highlight the practical interest of this strategy.