Pingtang, China – Nestled among the mountains in southwest China, the world's largest radio telescope signals Beijing's ambitions as a global center for scientific research. The Five-hundred-meter Aperture Spherical Telescope (FAST) -- the only significant instrument of its kind after the collapse of another telescope in Puerto Rico this month -- is about to open its doors for foreign astronomers to use, hoping to attract the world's top scientific talent. The world's second-largest radio telescope, at the Arecibo Observatory in Puerto Rico, was destroyed when its suspended 900-ton receiver platform came loose and plunged 140 meters onto the radio dish below. Wang Qiming, chief inspector of FAST's operations and development center, said during a rare visit by the foreign press last week that he had visited Arecibo. "We drew a lot of inspiration from its structure, which we gradually improved to build our telescope," he said.

As a simple and efficient optimization method in deep learning, stochastic gradient descent (SGD) has attracted tremendous attention. In the vanishing learning rate regime, SGD is now relatively well understood, and the majority of theoretical approaches to SGD set their assumptions in the continuous-time limit. However, the continuous-time predictions are unlikely to reflect the experimental observations well because the practice often runs in the large learning rate regime, where the training is faster and the generalization of models are often better. In this paper, we propose to study the basic properties of SGD and its variants in the non-vanishing learning rate regime. The focus is on deriving exactly solvable results and relating them to experimental observations. The main contributions of this work are to derive the stable distribution for discrete-time SGD in a quadratic loss function with and without momentum. Examples of applications of the proposed theory considered in this work include the approximation error of variants of SGD, the effect of mini-batch noise, the escape rate from a sharp minimum, and and the stationary distribution of a few second order methods.

In the problem of learning a class ratio from unlabeled data, which we call CR learning, the training data is unlabeled, and only the ratios, or proportions, of examples receiving each label are given. The goal is to learn a hypothesis that predicts the proportions of labels on the distribution underlying the sample. This model of learning is applicable to a wide variety of settings, including predicting the number of votes for candidates in political elections from polls. In this paper, we formally define this class and resolve foundational questions regarding the computational complexity of CR learning and characterize its relationship to PAC learning. Among our results, we show, perhaps surprisingly, that for finite VC classes what can be efficiently CR learned is a strict subset of what can be learned efficiently in PAC, under standard complexity assumptions. We also show that there exist classes of functions whose CR learnability is independent of ZFC, the standard set theoretic axioms. This implies that CR learning cannot be easily characterized (like PAC by VC dimension).

Electroencephalography is frequently used for diagnostic evaluation of various brain-related disorders due to its excellent resolution, non-invasive nature and low cost. However, manual analysis of EEG signals could be strenuous and a time-consuming process for experts. It requires long training time for physicians to develop expertise in it and additionally experts have low inter-rater agreement (IRA) among themselves. Therefore, many Computer Aided Diagnostic (CAD) based studies have considered the automation of interpreting EEG signals to alleviate the workload and support the final diagnosis. In this paper, we present an automatic binary classification framework for brain signals in multichannel EEG recordings. We propose to use Wavelet Packet Decomposition (WPD) techniques to decompose the EEG signals into frequency sub-bands and extract a set of statistical features from each of the selected coefficients. Moreover, we propose a novel method to reduce the dimension of the feature space without compromising the quality of the extracted features. The extracted features are classified using different Gradient Boosting Decision Tree (GBDT) based classification frameworks, which are CatBoost, XGBoost and LightGBM. We used Temple University Hospital EEG Abnormal Corpus V2.0.0 to test our proposed technique. We found that CatBoost classifier achieves the binary classification accuracy of 87.68%, and outperforms state-of-the-art techniques on the same dataset by more than 1% in accuracy and more than 3% in sensitivity. The obtained results in this research provide important insights into the usefulness of WPD feature extraction and GBDT classifiers for EEG classification.

It has been insufficiently explored how to perform density-based clustering by exploiting textual attributes on social media. In this paper, we aim at discovering a social point-of-interest (POI) boundary, formed as a convex polygon. More specifically, we present a new approach and algorithm, built upon our earlier work on social POI boundary estimation (SoBEst). This SoBEst approach takes into account both relevant and irrelevant records within a geographic area, where relevant records contain a POI name or its variations in their text field. Our study is motivated by the following empirical observation: a fixed representative coordinate of each POI that SoBEst basically assumes may be far away from the centroid of the estimated social POI boundary for certain POIs. Thus, using SoBEst in such cases may possibly result in unsatisfactory performance on the boundary estimation quality (BEQ), which is expressed as a function of the $F$-measure. To solve this problem, we formulate a joint optimization problem of simultaneously finding the radius of a circle and the POI's representative coordinate $c$ by allowing to update $c$. Subsequently, we design an iterative SoBEst (I-SoBEst) algorithm, which enables us to achieve a higher degree of BEQ for some POIs. The computational complexity of the proposed I-SoBEst algorithm is shown to scale linearly with the number of records. We demonstrate the superiority of our algorithm over competing clustering methods including the original SoBEst.

Increased availability of epidemiological data, novel digital data streams, and the rise of powerful machine learning approaches have generated a surge of research activity on real-time epidemic forecast systems. In this paper, we propose the use of a novel data source, namely retail market data to improve seasonal influenza forecasting. Specifically, we consider supermarket retail data as a proxy signal for influenza, through the identification of sentinel baskets, i.e., products bought together by a population of selected customers. We develop a nowcasting and forecasting framework that provides estimates for influenza incidence in Italy up to 4 weeks ahead. We make use of the Support Vector Regression (SVR) model to produce the predictions of seasonal flu incidence. Our predictions outperform both a baseline autoregressive model and a second baseline based on product purchases. The results show quantitatively the value of incorporating retail market data in forecasting models, acting as a proxy that can be used for the real-time analysis of epidemics.

Level Set Estimation (LSE) is an important problem with applications in various fields such as material design, biotechnology, machine operational testing, etc. Existing techniques suffer from the scalability issue, that is, these methods do not work well with high dimensional inputs. This paper proposes novel methods to solve the high dimensional LSE problems using Bayesian Neural Networks. In particular, we consider two types of LSE problems: (1) \textit{explicit} LSE problem where the threshold level is a fixed user-specified value, and, (2) \textit{implicit} LSE problem where the threshold level is defined as a percentage of the (unknown) maximum of the objective function. For each problem, we derive the corresponding theoretic information based acquisition function to sample the data points so as to maximally increase the level set accuracy. Furthermore, we also analyse the theoretical time complexity of our proposed acquisition functions, and suggest a practical methodology to efficiently tune the network hyper-parameters to achieve high model accuracy. Numerical experiments on both synthetic and real-world datasets show that our proposed method can achieve better results compared to existing state-of-the-art approaches.

Continual learning (CL) aims to enable information systems to learn from a continuous data stream across time. However, it is difficult for existing deep learning architectures to learn a new task without largely forgetting previously acquired knowledge. Furthermore, CL is particularly challenging for language learning, as natural language is ambiguous: it is discrete, compositional, and its meaning is context-dependent. In this work, we look at the problem of CL through the lens of various NLP tasks. Our survey discusses major challenges in CL and current methods applied in neural network models. We also provide a critical review of the existing CL evaluation methods and datasets in NLP.

In the present paper we use a range of modeling techniques to investigate whether an abstract phone could emerge from exposure to speech sounds. We test two opposing principles regarding the development of language knowledge in linguistically untrained language users: Memory-Based Learning (MBL) and Error-Correction Learning (ECL). A process of generalization underlies the abstractions linguists operate with, and we probed whether MBL and ECL could give rise to a type of language knowledge that resembles linguistic abstractions. Each model was presented with a significant amount of pre-processed speech produced by one speaker. We assessed the consistency or stability of what the models have learned and their ability to give rise to abstract categories. Both types of models fare differently with regard to these tests. We show that ECL learning models can learn abstractions and that at least part of the phone inventory can be reliably identified from the input.

But there is no doubt that the pandemic has hastened the adoption of emerging digital technologies, ushered in a new era of remote and flexible working arrangements, increased organisations' reliance on digital infrastructure and exposed our tech-related strengths and weaknesses alike. Leaving 2020 in the rear-view mirror, we count down our top 10 predictions for 2021 and beyond in the domain of Digital Law in Australia. Despite an existing principles-based framework for the protection of privacy under the Privacy Act, in recent years the Federal Government has preferred to introduce parallel privacy requirements, such as the 13 Privacy Safeguards under the Consumer Data Right legislation and the privacy aspects of the upcoming Data Availability and Transparency Act for Government agencies. These nascent regimes are similar enough to the existing privacy regime to encourage complacency and different enough to give any compliance function a headache. Overlapping and often sectorial regulation adds to the increasing complexity of privacy law in Australia.