The South Australian Government had recently launched a free autonomous vehicle trial that will deliver a first and last mile service between the Playford Alive Township and the Munno Para Train Station. As reported, the autonomous electric bus will run approximately every half hour on weekdays and will carry passengers from the Munno Para Railway Station to the Playford Alive Township, including stops at Mark Oliphant College and the Stretton Centre. The free autonomous shuttle bus service will include a morning and afternoon service window, with passengers collected from designated pick up points on the route. The Munno Para service is phase two of the Playford Connect autonomous bus trial. It follows phase one, which provided a free park and ride service from the Lionsgate Carpark through to the Lyell McEwin Hospital in Elizabeth Vale from October 2018 to June 2019.

How can we trust the networks?" They answered: "We know that a network is quite reliable when its inputs come from its training set. But these critical systems will have inputs corresponding to new, often unanticipated situations. There are numerous examples where a network gives poor responses for untrained inputs." David Lorge Parnas followed up on this discussion in his Letter to the Editor (Feb. We wish to point out that machine learning-based systems, including commercial ones performing safety critical tasks, can fail not only under "unanticipated situations" (noted by Lewis and Denning) or "when it encounters data radically different from its training set" (noted by Parnas), but also under normal situations, even on data that is extremely similar to its training set. The Apollo self-driving team confirmed "it might happen" because the system was "deep learning trained." Now, after a further investigation, we have found that in 24 of these 27 failed tests, the 10 random points ...

Bayesian synthetic likelihood (BSL) is a popular method for estimating the parameter posterior distribution for complex statistical models and stochastic processes that possess a computationally intractable likelihood function. Instead of evaluating the likelihood, BSL approximates the likelihood of a judiciously chosen summary statistic of the data via model simulation and density estimation. Compared to alternative methods such as approximate Bayesian computation (ABC), BSL requires little tuning and requires less model simulations than ABC when the chosen summary statistic is high-dimensional. The original synthetic likelihood relies on a multivariate normal approximation of the intractable likelihood, where the mean and covariance are estimated by simulation. An extension of BSL considers replacing the sample covariance with a penalised covariance estimator to reduce the number of required model simulations. Further, a semi-parametric approach has been developed to relax the normality assumption. In this paper, we present an R package called BSL that amalgamates the aforementioned methods and more into a single, easy-to-use and coherent piece of software. The R package also includes several examples to illustrate how to use the package and demonstrate the utility of the methods.

We propose a novel neural topic model in the Wasserstein autoencoders (WAE) framework. Unlike existing variational autoencoder based models, we directly enforce Dirichlet prior on the latent document-topic vectors. We exploit the structure of the latent space and apply a suitable kernel in minimizing the Maximum Mean Discrepancy (MMD) to perform distribution matching. We discover that MMD performs much better than the Generative Adversarial Network (GAN) in matching high dimensional Dirichlet distribution. We further discover that incorporating randomness in the encoder output during training leads to significantly more coherent topics. To measure the diversity of the produced topics, we propose a simple topic uniqueness metric. Together with the widely used coherence measure NPMI, we offer a more wholistic evaluation of topic quality. Experiments on several real datasets show that our model produces significantly better topics than existing topic models.

Are neural networks biased toward simple functions? Does depth always help learn more complex features? Is training the last layer of a network as good as training all layers? These questions seem unrelated at face value, but in this work we give all of them a common treatment from the spectral perspective. We will study the spectra of the *Conjugate Kernel*, CK, (also called the *Neural Network-Gaussian Process Kernel*), and the *Neural Tangent Kernel*, NTK. Roughly, the CK and the NTK tell us respectively "what a network looks like at initialization"and "what a network looks like during and after training." Their spectra then encode valuable information about the initial distribution and the training and generalization properties of neural networks. By analyzing the eigenvalues, we lend novel insights into the questions put forth at the beginning, and we verify these insights by extensive experiments of neural networks. We believe the computational tools we develop here for analyzing the spectra of CK and NTK serve as a solid foundation for future studies of deep neural networks. We have open-sourced the code for it and for generating the plots in this paper at github.com/thegregyang/NNspectra.

Applications that generate huge amounts of data in the form of fast streams are becoming increasingly prevalent, being therefore necessary to learn in an online manner. These conditions usually impose memory and processing time restrictions, and they often turn into evolving environments where a change may affect the input data distribution. Such a change causes that predictive models trained over these stream data become obsolete and do not adapt suitably to new distributions. Specially in these non-stationary scenarios, there is a pressing need for new algorithms that adapt to these changes as fast as possible, while maintaining good performance scores. Unfortunately, most off-the-shelf classification models need to be retrained if they are used in changing environments, and fail to scale properly. Spiking Neural Networks have revealed themselves as one of the most successful approaches to model the behavior and learning potential of the brain, and exploit them to undertake practical online learning tasks. Besides, some specific flavors of Spiking Neural Networks can overcome the necessity of retraining after a drift occurs. This work intends to merge both fields by serving as a comprehensive overview, motivating further developments that embrace Spiking Neural Networks for online learning scenarios, and being a friendly entry point for non-experts.

One of the most promising techniques in stream learning is the Spiking Neural Network, and some of them use an interesting population encoding scheme to transform the incoming stimuli into spikes. This study sheds lights on the key issue of this encoding scheme, the Gaussian receptive fields, and focuses on applying them as a pre-processing technique to any dataset in order to gain representativeness, and to boost the predictive performance of the stream learning methods. Experiments with synthetic and real data sets are presented, and lead to confirm that our approach can be applied successfully as a general pre-processing technique in many real cases. Keywords: Stream learning, gaussian receptive fields, population encoding, spiking neural networks 1. Introduction The continuous production of tremendous amount of data in the form of fast streams upsets the traditional view in machine learning, thus giving rise to a new emerging paradigm called stream learning (SL). These streams of data evolve generally over time and may be occasionally affected by a change (concept drift) which impacts on their input data distribution, without following the fundamental hypothesis of stationarity upon which the learning theory is based. Learning in non-stationary environments has attracted much attention in the SL community in Corresponding author: jesus.lopez@tecnalia.com

Graph Neural Networks (GNNs) have become a topic of intense research recently due to their powerful capability in high-dimensional classification and regression tasks for graph-structured data. However, as GNNs typically define the graph convolution by the orthonormal basis for the graph Laplacian, they suffer from high computational cost when the graph size is large. This paper introduces the Haar basis, a sparse and localized orthonormal system for graph, constructed from a coarse-grained chain on the graph. The graph convolution under Haar basis --- the Haar convolution can be defined accordingly for GNNs. The sparsity and locality of the Haar basis allow Fast Haar Transforms (FHTs) on graph, by which a fast evaluation of Haar convolution between the graph signals and the filters can be achieved. We conduct preliminary experiments on GNNs equipped with Haar convolution, which can obtain state-of-the-art results for a variety of geometric deep learning tasks.

Domino's AI pizza checker boosts customer quality scores 15% in a month Domino's Australia said its AI/machine learning order accuracy tool -- the so-called "DOM Pizza Checker" -- is turning in some pretty amazing results, just one month after it was first introduced this year and the rest of the Domino's system is watching closely. Natural Adversarial Examples Researchers introduce natural adversarial examples -- real-world, unmodified, and naturally occurring examples that cause classifier accuracy to significantly degrade. They curate 7,500 natural adversarial examples and release them in an ImageNet classifier test set that they call IMAGENET-A.

Batch-normalization (BN) layers are thought to be an integrally important layer type in today's state-of-the-art deep convolutional neural networks for computer vision tasks such as classification and detection. However, BN layers introduce complexity and computational overheads that are highly undesirable for training and/or inference on low-power custom hardware implementations of real-time embedded vision systems such as UAVs, robots and Internet of Things (IoT) devices. They are also problematic when batch sizes need to be very small during training, and innovations such as residual connections introduced more recently than BN layers could potentially have lessened their impact. In this paper we aim to quantify the benefits BN layers offer in image classification networks, in comparison with alternative choices. In particular, we study networks that use shifted-ReLU layers instead of BN layers. We found, following experiments with wide residual networks applied to the ImageNet, CIFAR 10 and CIFAR 100 image classification datasets, that BN layers do not consistently offer a significant advantage. We found that the accuracy margin offered by BN layers depends on the data set, the network size, and the bit-depth of weights. We conclude that in situations where BN layers are undesirable due to speed, memory or complexity costs, that using shifted-ReLU layers instead should be considered; we found they can offer advantages in all these areas, and often do not impose a significant accuracy cost.