Smartphones, wearables, and Internet of Things (IoT) devices produce a wealth of data that cannot be accumulated in a centralized repository for learning supervised models due to privacy, bandwidth limitations, and the prohibitive cost of annotations. Federated learning provides a compelling framework for learning models from decentralized data, but conventionally, it assumes the availability of labeled samples, whereas on-device data are generally either unlabeled or cannot be annotated readily through user interaction. To address these issues, we propose a self-supervised approach termed \textit{scalogram-signal correspondence learning} based on wavelet transform to learn useful representations from unlabeled sensor inputs, such as electroencephalography, blood volume pulse, accelerometer, and WiFi channel state information. Our auxiliary task requires a deep temporal neural network to determine if a given pair of a signal and its complementary viewpoint (i.e., a scalogram generated with a wavelet transform) align with each other or not through optimizing a contrastive objective. We extensively assess the quality of learned features with our multi-view strategy on diverse public datasets, achieving strong performance in all domains. We demonstrate the effectiveness of representations learned from an unlabeled input collection on downstream tasks with training a linear classifier over pretrained network, usefulness in low-data regime, transfer learning, and cross-validation. Our methodology achieves competitive performance with fully-supervised networks, and it outperforms pre-training with autoencoders in both central and federated contexts. Notably, it improves the generalization in a semi-supervised setting as it reduces the volume of labeled data required through leveraging self-supervised learning.

A wing-flapping drone inspired by the world's fastest bird has been developed that could one day find use in everything from surveillance operations to flower pollination. An international team of researchers designed the 26-gram ornithopter drone to hover, dart, glide, and dive just like a swift, making it far more versatile than a traditional quadcopter drone. "Unlike common quadcopters that are quite intrusive and not very agile, biologically-inspired drones could be used very successfully in a range of environments," said Dr Yao-Wei Chin, a research scientist from the National University of Singapore who led the project. "The light weight and the slow beating wings of the ornithopter poses less danger to the public that quadcopter drones in the event of a crash and given sufficient thrust and power banks it could be modified to carry different payloads depending on what is required." The researchers expect the first commercial use of the drone could be in monitoring large crowds or inspecting crops in fields.

What might self-driving cars do for people with a disability in places like the Central Coast? Notwithstanding recent improvements, the Coast still has a relatively poor public transport network, something that hinders people who cannot drive themselves to work or social engagements. And so it was with much anticipation I travelled to the Newcastle foreshore to experience for myself the driverless shuttle imported from France. If the shuttle was to be trialled and accepted here on the Central Coast, I could see it being a huge advantage to people living with disabilities, mobility issues, the elderly and people who are unable to drive. In 2016 approximately 6.4 per cent of people on the Central Coast needed help in their day-to-day lives due to disability.

Huge fleets of Chinese fishing boats have been caught stealthily operating in North Korean waters – while having their tracking systems turned off. The potentially illegal fishing operation was revealed through a combination of artificial intelligence, radar and satellite data. A study published today in the journal Science Advances details how more than 900 vessels of Chinese origin (over 900 in 2017 and over 700 in 2018) likely caught more than 160,000 metric tons --close to half a billion dollars' worth -- of Pacific flying squid over two years. This may be in violation of United Nations sanctions, which began restricting North Korea from foreign fishing in September 2017 following the country's ballistic missile tests. Illegal fishing threatens fish stocks and maritime ecosystem, and can also jeopardise food security for legitimate fishers.

Decision tree learning is a widely used approach in machine learning, favoured in applications that require concise and interpretable models. Heuristic methods are traditionally used to quickly produce models with reasonably high accuracy. A commonly criticised point, however, is that the resulting trees may not necessarily be the best representation of the data in terms of accuracy, size, and other considerations such as fairness. In recent years, this motivated the development of optimal classification tree algorithms that globally optimise the decision tree in contrast to heuristic methods that perform a sequence of locally optimal decisions. We follow this line of work and provide a novel algorithm for learning optimal classification trees based on dynamic programming and search. Our algorithm supports constraints on the depth of the tree and number of nodes and we argue it can be extended with other requirements. The success of our approach is attributed to a series of specialised techniques that exploit properties unique to classification trees. Whereas algorithms for optimal classification trees have traditionally been plagued by high runtimes and limited scalability, we show in a detailed experimental study that our approach uses only a fraction of the time required by the state-of-the-art and can handle datasets with tens of thousands of instances, providing several orders of magnitude improvements and notably contributing towards the practical realisation of optimal decision trees.

The design and implementation of an e-voting system is a challenging task. Formal analysis can be of great help here. In particular, it can lead to a better understanding of how the voting system works, and what requirements on the system are relevant. In this paper, we propose that the state-of-art model checker Uppaal provides a good environment for modelling and preliminary verification of voting protocols. To illustrate this, we present an Uppaal model of Pr\^et \`a Voter, together with some natural extensions. We also show how to verify a variant of receipt-freeness, despite the severe limitations of the property specification language in the model checker.

Distributional reinforcement learning (RL) has achieved state-of-the-art performance in Atari games by recasting the traditional RL into a distribution estimation problem, explicitly estimating the probability distribution instead of the expectation of a total return. The bottleneck in distributional RL lies in the estimation of this distribution where one must resort to an approximate representation of the return distributions which are infinite-dimensional. Most existing methods focus on learning a set of predefined statistic functionals of the return distributions requiring involved projections to maintain the order statistics. We take a different perspective using deterministic sampling wherein we approximate the return distributions with a set of deterministic particles that are not attached to any predefined statistic functional, allowing us to freely approximate the return distributions. The learning is then interpreted as evolution of these particles so that a distance between the return distribution and its target distribution is minimized. This learning aim is realized via maximum mean discrepancy (MMD) distance which in turn leads to a simpler loss amenable to backpropagation. Experiments on the suite of Atari 2600 games show that our algorithm outperforms the standard distributional RL baselines and sets a new record in the Atari games for non-distributed agents.

Recent work has demonstrated how data-driven AI methods can leverage consumer protection by supporting the automated analysis of legal documents. However, a shortcoming of data-driven approaches is poor explainability. We posit that in this domain useful explanations of classifier outcomes can be provided by resorting to legal rationales. We thus consider several configurations of memory-augmented neural networks where rationales are given a special role in the modeling of context knowledge. Our results show that rationales not only contribute to improve the classification accuracy, but are also able to offer meaningful, natural language explanations of otherwise opaque classifier outcomes.

Multi-view clustering integrates multiple feature sets, which reveal distinct aspects of the data and provide complementary information to each other, to improve the clustering performance. It remains challenging to effectively exploit complementary information across multiple views since the original data often contain noise and are highly redundant. Moreover, most existing multi-view clustering methods only aim to explore the consistency of all views while ignoring the local structure of each view. However, it is necessary to take the local structure of each view into consideration, because different views would present different geometric structures while admitting the same cluster structure. To address the above issues, we propose a novel multi-view subspace clustering method via simultaneously assigning weights for different features and capturing local information of data in view-specific self-representation feature spaces. Especially, a common cluster structure regularization is adopted to guarantee consistency among different views. An efficient algorithm based on an augmented Lagrangian multiplier is also developed to solve the associated optimization problem. Experiments conducted on several benchmark datasets demonstrate that the proposed method achieves state-of-the-art performance. We provide the Matlab code on https://github.com/Ekin102003/JFLMSC.

Conformal methods create prediction bands that control average coverage assuming solely i.i.d. data. Although the literature has mostly focused on prediction intervals, more general regions can often better represent uncertainty. For instance, a bimodal target is better represented by the union of two intervals. Such prediction regions are obtained by CD-split, which combines the split method and a data-driven partition of the feature space which scales to high dimensions. In this paper, we provide new theoretical properties and simulations related to CD-split. We show that CD-split converges asymptotically to the oracle highest density set. In particular, we show that CD-split satisfies local and asymptotic conditional validity. We also present many new simulations, which show how to tune CD-split and compare it to other methods in the literature. In a wide variety of these simulations, CD-split has a better conditional coverage and yields smaller prediction regions than other methods.