Despite the recent success of stochastic gradient descent in deep learning, it is often difficult to train a deep neural network with an inappropriate choice of its initial parameters. Even if training is successful, it has been known that the initial parameter configuration may negatively impact generalization. In this paper, we propose an unsupervised algorithm to find good initialization for input data, given that a downstream task is d-way classification. We then conjecture that the success of learning is directly related to how diverse downstream tasks are in the vicinity of the initial parameters. We thus design an algorithm that encourages small perturbation to the initial parameter configuration leads to a diverse set of d-way classification tasks. In other words, the proposed algorithm ensures a solution to any downstream task to be near the initial parameter configuration. We empirically evaluate the proposed algorithm on various tasks derived from MNIST with a fully connected network. In these experiments, we observe that our algorithm improves average test accuracy across most of these tasks, and that such improvement is greater when the number of labelled examples is small.

Out-of-distribution (OoD) detection for data-based programs is a goal of paramount importance. Common approaches in the literature tend to train detectors requiring inside-of-distribution (in-distribution, or IoD) and OoD validation samples, and/or implement confidence metrics that are often abstract and therefore difficult to interpret. In this work, we propose to use existing work from the field of explainable AI, namely the PARTICUL pattern identification algorithm, in order to build more interpretable and robust OoD detectors for visual classifiers. Crucially, this approach does not require to retrain the classifier and is tuned directly to the IoD dataset, making it applicable to domains where OoD does not have a clear definition. Moreover, pattern identification allows us to provide images from the IoD dataset as reference points to better explain the confidence scores. We demonstrates that the detection capabilities of this approach are on par with existing methods through an extensive benchmark across four datasets and two definitions of OoD. In particular, we introduce a new benchmark based on perturbations of the IoD dataset which provides a known and quantifiable evaluation of the discrepancy between the IoD and OoD datasets that serves as a reference value for the comparison between various OoD detection methods. Our experiments show that the robustness of all metrics under test does not solely depend on the nature of the IoD dataset or the OoD definition, but also on the architecture of the classifier, which stresses the need for thorough experimentations for future work on OoD detection.

Researchers in Japan and The Netherlands have, for the first time, used machine learning techniques, in particular artificial neural networks (ANNs), to predict the Indian Ocean Dipole (IOD), a positive phase of which has affected weather and climate in India and Australia in a spectacular fashion so far in 2019-20. The IOD has both positive and negative phases, and signals large socio-economic impacts on many countries and hence predicting the IOD well in advance will benefit the affected societies, note authors JV Ratnam and Swadhin K Behera (Application Laboratory, Japan Agency for Marine-Earth Science and Technology, Yokohama) and HA Dijkstra (Institute for Marine and Atmospheric Research Utrecht, Utrecht University in The Netherlands) in a paper published by Nature. The IOD is a mode of climate variability observed in the Indian Ocean sea surface temperature anomalies with one pole in Sumatra (Indonesia) and the other near East Africa. Therefore, the IOD is represented by an index derived from the gradient between the western equatorial Indian Ocean and the south-eastern equatorial Indian Ocean. It starts sometime in May-June, peaks in September-October and ends in November (2019's rather strong positive phase of the IOD lasted into early January of 2020).