This is an Inside Science story. When a song plays on the radio, there are invisible forces at work that go beyond the creative scope of the writing, performing and producing of the song. One of those ineffable qualities is audio mastering, a process that smooths out the song and optimizes the listening experience on any device. Now, artificial intelligence algorithms are starting to work their way into this undertaking. "Mastering is a bit of a black art," explained Thomas Birtchnell, a researcher at the University of Wollongong in Australia.

Over the past years, distributed semantic representations have proved to be effective and flexible keepers of prior knowledge to be integrated into downstream applications. This survey focuses on the representation of meaning. We start from the theoretical background behind word vector space models and highlight one of their major limitations: the meaning conflation deficiency, which arises from representing a word with all its possible meanings as a single vector. Then, we explain how this deficiency can be addressed through a transition from the word level to the more fine-grained level of word senses (in its broader acceptation) as a method for modelling unambiguous lexical meaning. We present a comprehensive overview of the wide range of techniques in the two main branches of sense representation, i.e., unsupervised and knowledge-based. Finally, this survey covers the main evaluation procedures and applications for this type of representation, and provides an analysis of four of its important aspects: interpretability, sense granularity, adaptability to different domains and compositionality.

The practice of transforming raw data to a feature space so that inference can be performed in that space has been popular for many years. Recently, rapid progress in deep neural networks has given both researchers and practitioners enhanced methods that increase the richness of feature representations, be it from images, text or speech. In this work we show how a constructed latent space can be explored in a controlled manner and argue that this complements well founded inference methods. For constructing the latent space a Variational Autoencoder is used. We present a novel controller module that allows for smooth traversal in the latent space and construct an end-to-end trainable framework. We explore the applicability of our method for performing spatial transformations as well as kinematics for predicting future latent vectors of a video sequence.

Neural networks can achieve extraordinary results on a wide variety of tasks. However, when they attempt to sequentially learn a number of tasks, they tend to learn the new task while destructively forgetting previous tasks. One solution to this problem is pseudo-rehearsal, which involves learning the new task while rehearsing generated items representative of previous task/s. We demonstrate that pairing pseudo-rehearsal methods with a generative network is an effective solution to this problem in reinforcement learning. Our method iteratively learns three Atari 2600 games while retaining above human level performance on all three games, performing similar to a network which rehearses real examples from all previously learnt tasks.

Node embedding is the task of extracting informative and descriptive features over the nodes of a graph. The importance of node embeddings for graph analytics, as well as learning tasks such as node classification, link prediction and community detection, has led to increased interest on the problem leading to a number of recent advances. Much like PCA in the feature domain, node embedding is an inherently \emph{unsupervised} task; in lack of metadata used for validation, practical methods may require standardization and limiting the use of tunable hyperparameters. Finally, node embedding methods are faced with maintaining scalability in the face of large-scale real-world graphs of ever-increasing sizes. In the present work, we propose an adaptive node embedding framework that adjusts the embedding process to a given underlying graph, in a fully unsupervised manner. To achieve this, we adopt the notion of a tunable node similarity matrix that assigns weights on paths of different length. The design of the multilength similarities ensures that the resulting embeddings also inherit interpretable spectral properties. The proposed model is carefully studied, interpreted, and numerically evaluated using stochastic block models. Moreover, an algorithmic scheme is proposed for training the model parameters effieciently and in an unsupervised manner. We perform extensive node classification, link prediction, and clustering experiments on many real world graphs from various domains, and compare with state-of-the-art scalable and unsupervised node embedding alternatives. The proposed method enjoys superior performance in many cases, while also yielding interpretable information on the underlying structure of the graph.

Artificial intelligence (AI) might be technology's Holy Grail, but Australia's Human Rights Commissioner Edward Santow has warned about the need for responsible innovation and an understanding of the challenges new technology poses for basic human rights. "AI is enabling breakthroughs right now: Healthcare, robotics, and manufacturing; pretty soon we're told AI will bring us everything from the perfect dating algorithm to interstellar travel -- it's easy in other words to get carried away, yet we should remember AI is still in its infancy," Santow told the Human Rights & Technology conference in Sydney in July. Santow was launching the Human Rights and Technology Issues Paper, which was described as the beginning of a major project by the Human Rights Commission to protect the rights of Australians in a new era of technological change. The paper [PDF] poses questions centred on what protections are needed when AI is used in decisions that affect the basic rights of people. It asks also what is required from lawmakers, governments, researchers, developers, and tech companies big and small. Pointing to Microsoft's AI Twitter bot Tay, which in March 2016 showed the ugly side of humanity -- at least as present on social media -- Santow said it is a key example of how AI must be right before it's unleashed onto humans.

Artificial intelligence (AI) – the science of making computers mimic humans using logic, decision trees, deep learning, and machine learning – is fast approaching the market opportunity around preventive and predictive maintenance. According to a recent GlobalData survey, the top two business challenges in Australia are in improving operational efficiency and reducing costs. Many businesses, such as manufacturers, producers of natural resources, through to the agriculture and health sectors, need ongoing reliability of machines and their constituent parts to keep the lights on in the business. Unplanned outages, for example, can cost an oil and gas company, on average $50 million dollars annually. In the case of a windfarm, in the event of one single fail, an entire turbine needs to come down, a technical crew with a crane needs to be on site costing $100,000 or more for each time a part fails.

The rigorous quantification of uncertainty in geophysical inversions is a challenging problem. Inversions are often ill-posed and the likelihood surface may be multi-modal; properties of any single mode become inadequate uncertainty measures, and sampling methods become inefficient for irregular posteriors or high-dimensional parameter spaces. We explore the influences of different choices made by the practitioner on the efficiency and accuracy of Bayesian geophysical inversion methods that rely on Markov chain Monte Carlo sampling to assess uncertainty, using a multi-sensor inversion of the three-dimensional structure and composition of a region in the Cooper Basin of South Australia as a case study. The inversion is performed using an updated version of the Obsidian distributed inversion software. We find that the posterior for this inversion has complex local covariance structure, hindering the efficiency of adaptive sampling methods that adjust the proposal based on the chain history. Within the context of a parallel-tempered Markov chain Monte Carlo scheme for exploring high-dimensional multi-modal posteriors, a preconditioned Crank-Nicholson proposal outperforms more conventional forms of random walk. Aspects of the problem setup, such as priors on petrophysics or on 3-D geological structure, affect the shape and separation of posterior modes, influencing sampling performance as well as the inversion results. Use of uninformative priors on sensor noise can improve inversion results by enabling optimal weighting among multiple sensors even if noise levels are uncertain. Efficiency could be further increased by using posterior gradient information within proposals, which Obsidian does not currently support, but which could be emulated using posterior surrogates.

The economic and social impact of poor air quality in towns and cities is increasingly being recognised, together with the need for effective ways of creating awareness of real-time air quality levels and their impact on human health. With local authority maintained monitoring stations being geographically sparse and the resultant datasets also featuring missing labels, computational data-driven mechanisms are needed to address the data sparsity challenge. In this paper, we propose a machine learning-based method to accurately predict the Air Quality Index (AQI), using environmental monitoring data together with meteorological measurements. To do so, we develop an air quality estimation framework that implements a neural network that is enhanced with a novel Non-linear Autoregressive neural network with exogenous input (NARX), especially designed for time series prediction. The framework is applied to a case study featuring different monitoring sites in London, with comparisons against other standard machine-learning based predictive algorithms showing the feasibility and robust performance of the proposed method for different kinds of areas within an urban region.

Deep learning models are known to be overconfident in their predictions on out of distribution inputs. There have been several pieces of work to address this issue, including a number of approaches for building Bayesian neural networks, as well as closely related work on detection of out of distribution samples. Recently, there has been work on building classifiers that are robust to out of distribution samples by adding a regularization term that maximizes the entropy of the classifier output on out of distribution data. To approximate out of distribution samples (which are not known apriori), a GAN was used for generation of samples at the edges of the training distribution. In this paper, we introduce an alternative GAN based approach for building a robust classifier, where the idea is to use the GAN to explicitly generate out of distribution samples that the classifier is confident on (low entropy), and have the classifier maximize the entropy for these samples.