Children have this incredible capability to transform their reality in a matter of seconds into something magical: a place where everything can happen and the only limits are one's imagination. No matter how old we are, if we try hard enough we still can remember bits and pieces of this land of beauty where legends came alive, where the lines of the carpet were windy mountain roads for matchbox cars and dolls had their own social life. If we look closely we will realize that Toy Story is rather boring, because our toys had way better lives and adventures. One major effect of growing up -- or side effect, some may argue -- is that we slowly let go of our vivid imagination. Fantasies become scenarios more anchored in our reality and we name them plans: future plans, business plans, and any other plan you can think of.

In October, the Institute announced the creation of the MIT Stephen A. Schwarzman College of Computing, an ambitious new enterprise that will allow students to better tailor their educational interests to their goals. But the ideas driving this exciting new effort may carry a distant echo -- especially among alumni were at MIT during the 1980s -- from the time leadership launched another computing enterprise that dramatically changed how undergraduates and graduate students learned. Project Athena was a campus-wide effort to make the tools of computing available to every discipline at the Institute and provide students with systematic access to computers. A new project that featured computer workstations and educational programming, Athena was a milestone in the history of distributed systems and inspired programs like Kerberos. It also revolutionized educational computing for the Institute and beyond, and created the computing environment that many students and faculty still work in today.

While Mongolia may be most sparsely populated independent country in the world, in the European Union that claim goes to Finland. With just 5.5 million people (that's about the same population as Houston and Chicago put together, just without that whole deadly crime thing) Finland has many claims to fame. She is a country of great natural beauty that has influenced generations of minimalist industrial designers. More importantly though, the country follows the Nordic model of capitalism and has thus became one of the few working examples of a progressive, socially sensitive state with superb welfare, education, and healthcare services. It's no surprise then that the country's liberal administration is keen to explore the possibilities offered by artificial intelligence.

We say a lot with our hands. We spread them wide to indicate size, stab the air for emphasis and reach out to draw people in. Waving our hands about when we speak makes us appear less robotic – and that's true for robots too.

Following the announcement earlier this week of Google Cloud's AI Hub and Kubeflow Pipelines tools, Rajen Sheth, director of product management for Cloud AI, has outlined how the technology giant is working to ensure that its AI work is ethical and fair. In a blog post earlier this week titled'steering the right course for AI', he outlined what is seen as the main industry challenges to be overcome in order to make AI not just a reality, but one that is for the net good of society. Engaging with each of these in turn, he first suggests that unfair, or confirmation bias must be tackled "on multiple fronts," starting with awareness. "To foster a wider understanding of the need for fairness in technologies like machine learning, we've created educational resources like ml-fairness.com Google is also encouraging thorough documentation "as a means to better understand what goes on inside a machine learning solution". Within Google this takes the form of'model cards': "a standardised format for describing the goals, assumptions, performance metrics, and even ethical considerations of a machine learning model." Embedded documentation tools from Google Cloud, like the Inclusive ML Guide, integrated throughout AutoML, and TensorFlow Model Analysis (TFMA) and the What-If Tool all help with this. "I'm proud of the steps we're taking, and I believe the knowledge and tools we're developing will go a long way towards making AI more fair," he said, before reiterating that this is an industry-wide problem to be tackled. "No single company can solve such a complex problem alone.

The paper proposes an on-line monitoring framework for continuous real-time safety/security in learning-based control systems (specifically application to a unmanned ground vehicle). We monitor validity of mappings from sensor inputs to actuator commands, controller-focused anomaly detection (CFAM), and from actuator commands to sensor inputs, system-focused anomaly detection (SFAM). CFAM is an image conditioned energy based generative adversarial network (EBGAN) in which the energy based discriminator distinguishes between proper and anomalous actuator commands. SFAM is based on an action condition video prediction framework to detect anomalies between predicted and observed temporal evolution of sensor data. We demonstrate the effectiveness of the approach on our autonomous ground vehicle for indoor environments and on Udacity dataset for outdoor environments.

This paper is concerned with the approximation of high-dimensional functions in a statistical learning setting, by empirical risk minimization over model classes of functions in tree-based tensor format. These are particular classes of rank-structured functions that can be seen as deep neural networks with a sparse architecture related to the tree and multilinear activation functions. For learning in a given model class, we exploit the fact that tree-based tensor formats are multilinear models and recast the problem of risk minimization over a nonlinear set into a succession of learning problems with linear models. Suitable changes of representation yield numerically stable learning problems and allow to exploit sparsity. For high-dimensional problems or when only a small data set is available, the selection of a good model class is a critical issue. For a given tree, the selection of the tuple of tree-based ranks that minimize the risk is a combinatorial problem. Here, we propose a rank adaptation strategy which provides in practice a good convergence of the risk as a function of the model class complexity. Finding a good tree is also a combinatorial problem, which can be related to the choice of a particular sparse architecture for deep neural networks. Here, we propose a stochastic algorithm for minimizing the complexity of the representation of a given function over a class of trees with a given arity, allowing changes in the topology of the tree. This tree optimization algorithm is then included in a learning scheme that successively adapts the tree and the corresponding tree-based ranks. Contrary to classical learning algorithms for nonlinear model classes, the proposed algorithms are numerically stable, reliable, and require only a low level expertise of the user.

Abstract--Matrix factorization (MF) has been widely used to discover the low-rank structure and to predict the missing entries of data matrix. In many real-world learning systems, the data matrix can be very high-dimensional but sparse. This poses an imbalanced learning problem, since the scale of missing entries is usually much larger than that of observed entries, but they cannot be ignored due to the valuable negative signal. For efficiency concern, existing work typically applies a uniform weight on missing entries to allow a fast learning algorithm. However, this simplification will decrease modeling fidelity, resulting in suboptimal performance for downstream applications. In this work, we weight the missing data non-uniformly, and more generically, we allow any weighting strategy on the missing data. To address the efficiency challenge, we propose a fast learning method, for which the time complexity is determined by the number of observed entries in the data matrix, rather than the matrix size. The key idea is twofold: 1) we apply truncated SVD on the weight matrix to get a more compact representation of the weights, and 2) we learn MF parameters with element-wise alternating least squares (eALS) and memorize the key intermediate variables to avoid repeating computations that are unnecessary. We conduct extensive experiments on two recommendation benchmarks, demonstrating the correctness, efficiency, and effectiveness of our fast eALS method. Atrices are a common data structure to represent the relation between two types of entities in learning systems [1]-[3]. In relational learning, matrix factorization (MF) is a popular approach for dimension reduction by representing the rows (entities of one type) and columns (entities of another type) as two low-rank matrices. The optimization of dimension reduction is usually achieved by minimizing the reconstruction error between the low-rank model and the original data. Xiangnan He and Tat-Seng Chua are with the School of Computing, National University of Singapore, Singapore, 117417. Jinhui Tang is with the School of Computer Science and Engineering, Nanjing University of Science and Technology, Nanjing, Jiangsu, China, 210094.

We investigate 1) the rate at which refined properties of the empirical risk---in particular, gradients---converge to their population counterparts in standard non-convex learning tasks, and 2) the consequences of this convergence for optimization. Our analysis follows the tradition of norm-based capacity control. We propose vector-valued Rademacher complexities as a simple, composable, and user-friendly tool to derive dimension-free uniform convergence bounds for gradients in non-convex learning problems. As an application of our techniques, we give a new analysis of batch gradient descent methods for non-convex generalized linear models and non-convex robust regression, showing how to use any algorithm that finds approximate stationary points to obtain optimal sample complexity, even when dimension is high or possibly infinite and multiple passes over the dataset are allowed. Moving to non-smooth models we show----in contrast to the smooth case---that even for a single ReLU it is not possible to obtain dimension-independent convergence rates for gradients in the worst case. On the positive side, it is still possible to obtain dimension-independent rates under a new type of distributional assumption.

Most of the datasets normally contain either numeric or categorical features. Mixed data comprises of both numeric and categorical features, and they frequently occur in various domains, such as health, finance, marketing, etc. Clustering is often sought on mixed data to find structures and to group similar objects. However, clustering mixed data is challenging because it is difficult to directly apply mathematical operations, such as summation, average etc. on the feature values of these datasets. In this paper, we review various types of mixed data clustering techniques in detail. We present a taxonomy to identify ten types of different mixed data clustering techniques. We also compare the performance of several mixed data clustering methods on publicly available datasets. The paper further identifies challenges in developing different mixed data clustering algorithms and provides guidelines for future directions in this area.