Google unveiled at CES on Wednesday a range of new ways to keep its Android devices connected, and that includes cars. As more vehicles go electric and automakers evolve into software developers, we can only expect to see more plays directed at turning cars into connected devices. One exemplar of this phenomenon is Volvo Cars, which will launch a direct integration with the Google Home ecosystem in the coming months, both Volvo and Google announced on Wednesday. The integration should allow car owners to turn their car on and off, control the temperature and get car information like battery life by issuing voice commands to Google Assistant-enabled home and mobile devices. Once customers pair their Volvo car to their Google account, they also can talk directly to Google while in their car.

This article is brought to you thanks to the collaboration of The European Sting with the World Economic Forum. A new survey has found that 60% of adults around the world expect that products and services using AI will profoundly change their daily life in the next 3-5 years. The same number also agree that AI products and services will make their life easier, but just half say they have more benefits than drawbacks. And, just 50% say they trust companies that use AI as much as they trust other companies. "In order to trust artificial intelligence, people must know and understand exactly what AI is, what it's doing, and its impact," said Kay Firth-Butterfield, Head of Artificial Intelligence and Machine Learning at the World Economic Forum.

The quantities of physical capital, or object counts, provide important insights into human activities and the socio-economic development of a region. For example, the number of buildings reflects the level of urbanization in a region; the number of brick kilns is related to the level of air pollution, and the number of cars correlates with the poverty level of a region. For example, the Demographic and Health Surveys (DHS) collects population-related statistics of about 90 countries at a cost of 1.9 million dollars over a five-year interval [1]. Recently, object detection in high-resolution satellite imagery has emerged as an alternative to ground-based survey data collection in socioeconomic monitoring tasks like counting brick kilns in Bangladesh [2] and counting solar panels in the U.S. [3]. A common detection-based pipeline [2, 4] to collect object count statistics over a large region exhaustively downloads all satellite images covering the target region, counts the objects in each image using a trained detection model, and takes the summation of counts in all the images to produce a total count.

Deep neural networks (DNNs) have achieved the state of the art performance in numerous fields. However, DNNs need high computation times, and people always expect better performance in a lower computation. Therefore, we study the human somatosensory system and design a neural network (SpinalNet) to achieve higher accuracy with fewer computations. Hidden layers in traditional NNs receive inputs in the previous layer, apply activation function, and then transfer the outcomes to the next layer. In the proposed SpinalNet, each layer is split into three splits: 1) input split, 2) intermediate split, and 3) output split. Input split of each layer receives a part of the inputs. The intermediate split of each layer receives outputs of the intermediate split of the previous layer and outputs of the input split of the current layer. The number of incoming weights becomes significantly lower than traditional DNNs. The SpinalNet can also be used as the fully connected or classification layer of DNN and supports both traditional learning and transfer learning. We observe significant error reductions with lower computational costs in most of the DNNs. Traditional learning on the VGG-5 network with SpinalNet classification layers provided the state-of-the-art (SOTA) performance on QMNIST, Kuzushiji-MNIST, EMNIST (Letters, Digits, and Balanced) datasets. Traditional learning with ImageNet pre-trained initial weights and SpinalNet classification layers provided the SOTA performance on STL-10, Fruits 360, Bird225, and Caltech-101 datasets. The scripts of the proposed SpinalNet are available at the following link: https://github.com/dipuk0506/SpinalNet

Clustering is a fundamental primitive in unsupervised learning which gives rise to a rich class of computationally-challenging inference tasks. In this work, we focus on the canonical task of clustering d-dimensional Gaussian mixtures with unknown (and possibly degenerate) covariance. Recent works (Ghosh et al. '20; Mao, Wein '21; Davis, Diaz, Wang '21) have established lower bounds against the class of low-degree polynomial methods and the sum-of-squares (SoS) hierarchy for recovering certain hidden structures planted in Gaussian clustering instances. Prior work on many similar inference tasks portends that such lower bounds strongly suggest the presence of an inherent statistical-to-computational gap for clustering, that is, a parameter regime where the clustering task is statistically possible but no polynomial-time algorithm succeeds. One special case of the clustering task we consider is equivalent to the problem of finding a planted hypercube vector in an otherwise random subspace. We show that, perhaps surprisingly, this particular clustering model does not exhibit a statistical-to-computational gap, even though the aforementioned low-degree and SoS lower bounds continue to apply in this case. To achieve this, we give a polynomial-time algorithm based on the Lenstra--Lenstra--Lovasz lattice basis reduction method which achieves the statistically-optimal sample complexity of d+1 samples. This result extends the class of problems whose conjectured statistical-to-computational gaps can be "closed" by "brittle" polynomial-time algorithms, highlighting the crucial but subtle role of noise in the onset of statistical-to-computational gaps.

Simulators are an essential tool for behavioural and interaction research on driving, due to the safety, cost, and experimental control issues of on-road driving experiments. The most advanced simulators use expensive 360 degree projections systems to ensure visual fidelity, full field of view, and immersion. However, similar visual fidelity can be achieved affordably using a virtual reality (VR) based visual interface. We present DReyeVR, an open-source VR based driving simulator platform designed with behavioural and interaction research priorities in mind. DReyeVR (read "driver") is based on Unreal Engine and the CARLA autonomous vehicle simulator and has features such as eye tracking, a functional driving heads-up display (HUD) and vehicle audio, custom definable routes and traffic scenarios, experimental logging, replay capabilities, and compatibility with ROS. We describe the hardware required to deploy this simulator for under $5000$ USD, much cheaper than commercially available simulators. Finally, we describe how DReyeVR may be leveraged to answer an interaction research question in an example scenario.

The human capability to reason about one domain by using knowledge of other domains has been researched for more than 50 years, but models that are formally sound and predict cognitive process are sparse. We propose a formally sound method that models associative reasoning by adapting logical reasoning mechanisms. In particular it is shown that the combination with large commensense knowledge within a single reasoning system demands for an efficient and powerful association technique. This approach is also used for modelling mind-wandering and the Remote Associates Test (RAT) for testing creativity. In a general discussion we show implications of the model for a broad variety of cognitive phenomena including consciousness.

The study of generalisation in deep Reinforcement Learning (RL) aims to produce RL algorithms whose policies generalise well to novel unseen situations at deployment time, avoiding overfitting to their training environments. Tackling this is vital if we are to deploy reinforcement learning algorithms in real world scenarios, where the environment will be diverse, dynamic and unpredictable. This survey is an overview of this nascent field. We provide a unifying formalism and terminology for discussing different generalisation problems, building upon previous works. We go on to categorise existing benchmarks for generalisation, as well as current methods for tackling the generalisation problem. Finally, we provide a critical discussion of the current state of the field, including recommendations for future work. Among other conclusions, we argue that taking a purely procedural content generation approach to benchmark design is not conducive to progress in generalisation, we suggest fast online adaptation and tackling RL-specific problems as some areas for future work on methods for generalisation, and we recommend building benchmarks in underexplored problem settings such as offline RL generalisation and reward-function variation.

The artificial intelligence (AI) that is already all around us cannot safely drive a car by himself, nor can they write compelling scripts. Still, every day the research to make them more capable gets new results, and in some cases, anyone with a computer and an Internet connection can help teach them something. Both Google and Microsoft have put online some experiments (interpretable as "games") with which to try to understand how to teach a computer to learn, or simply to get an idea of how smart some AI systems that already exist are. Google's can be found on the A.I. Experiments platform, while Microsoft's were mostly created by the Microsoft Garage research community. They are, of course, not the first such experiments.

As a university student, Abigail Annkah fell in love with mathematics, which soon led to her interest in artificial intelligence. After graduating from the African Institute for Mathematical Sciences, Abigail made it through the competitive process to become an AI resident at Google Research, Accra. After her residency, Google offered her a job and she’s now a research software engineer working on several high-profile projects.As Google grows its presence in Accra, we spoke to Abigail about the mapping project that motivates her, starting a new job while becoming a mother and the importance of inspiring young girls to enter STEM careers.How did your science background lead you to Google?I did my undergraduate degree in Bachelor of Science Statistics at the University of Ghana, finishing with a combined major in Mathematics and Statistics. During the second year of study, I stumbled upon Computational Maths, leading to my first taste of coding. I started taking extra credit courses, which really kickstarted my entry into AI. Then I joined the first cohort of the African Masters of Machine Intelligence program at African Institute for Mathematical Sciences with the support of Google and Facebook. The program intends to bridge the AI education gap in Africa and strengthen the growing data science ecosystem in the region — this was my first exposure to the world of Machine Learning.