CHICAGO & MADISON, Wis.--(BUSINESS WIRE)--Arturo, an artificial intelligence (AI) and deep-learning property analytics start-up, announced today the close of Seed Series Transaction, with lead investor and Arturo co-developer American Family Insurance. Following the close of the transaction, Madison-based American Family Insurance will remain a customer as well as an investor in Arturo. Arturo grew out of investment and more than three years of research by American Family Insurance on the application of artificial intelligence (AI) and deep learning to satellite, aerial, drone, and ground-level imagery to accurately assess physical property characteristics for residential and commercial properties. Arturo's AI-powered analytics generate detailed property information often in under five seconds. It enables a variety of businesses that insure, lend, invest, or manage residential or commercial properties to make more informed decisions and better manage risk with the most up to date information available.

Deep-learning algorithms have already mastered games like Starcraft to beat humans, and now they have shown they can team up to beat us too. The news: In a paper published in Science yesterday, DeepMind showed how it had let AI programs loose in a modified version of the 3D first-person video game Quake III Arena. The team used an algorithm called "For the Win," which trains a host of agents in parallel using reinforcement learning, the technique that lets AI learn which tactics work and which do not (and that famously enabled DeepMind's AI to win at Go). This time, AI agents were trained on around 450,000 games of Capture the Flag, the classic game that involves snatching a flag from your opponent's base while protecting your own. Each agent could only see a first-person view of the maze-like structure, just as a human player would.

Earlier this month Google quietly unveiled an incredibly unique opportunity for seasoned developers to explore pilot experiments based on some of Google's frontier AI research, aptly called "AI Workshop." Google already makes a wealth of AI research available on platforms from GitHub to its own AI Hub, complete with a searchable library of ready-to-use code examples, demonstrations and even wrappers around production systems. What makes AI Workshop so different from these other mediums is that it presents an early glimpse at selections from Google's bleeding edge enterprise AI research that might become future product offerings, allowing the research and developer community to provide feedback that can help influence those innovations, granting a rare opportunity to help shape the future of AI in the enterprise. The rise of deep learning has represented a unique era of collaboration between the commercial and research sectors. Many of the underlying toolkits, workflows, algorithms and even research models have all been released under open source licenses, with companies, academics and citizen researchers collaborating together to create innovative new applications and to improve the underlying infrastructure powering the modern deep learning revolution.

Few games are simpler in principle than capturing the flag (excepting perhaps tag or kick the can). Two teams each have a marker located at their respective bases, and the objective is to capture the other team's marker and return it safely back to their base. What's easily understood by humans is not quite so quickly grasped by machines, though. Where capture the flag is concerned in the video game domain, non-player characters have traditionally been programmed with heuristics and rules affording limited freedom in choice. But AI and machine learning promise to turn this paradigm on its head. In a paper published this week in the journal Science roughly a year following the preprint, researchers at DeepMind, the London-based subsidiary of Google parent company Alphabet, describe a system capable not only of learning how to play capture the flag in Id Software's Quake III Arena, but of devising entirely novel human-level team-based strategies.

New research from Google shows how machine learning could one day be used to detect signs of lung cancer earlier than often occurs today. Early warning: Danial Tse, a researcher at Google, developed an algorithm that beat a number of trained radiologists in testing. Tse and colleagues trained a deep-learning algorithm to detect malignant lung nodules in more than 42,000 CT scans. The resulting algorithms turned up 11% fewer false positives and 5% fewer false negatives than their human counterparts. The work is described in a paper published in the journal Nature today.

We propose a new approach, called cooperative neural networks (CoNN), which uses a set of cooperatively trained neural networks to capture latent representations that exploit prior given independence structure. The model is more flexible than traditional graphical models based on exponential family distributions, but incorporates more domain specific prior structure than traditional deep networks or variational autoencoders. The framework is very general and can be used to exploit the independence structure of any graphical model. We illustrate the technique by showing that we can transfer the independence structure of the popular Latent Dirichlet Allocation (LDA) model to a cooperative neural network, CoNN-sLDA. Empirical evaluation of CoNN-sLDA on supervised text classification tasks demonstrates that the theoretical advantages of prior independence structure can be realized in practice -we demonstrate a 23\% reduction in error on the challenging MultiSent data set compared to state-of-the-art.

Neural network learning is typically slow since backpropagation needs to compute full gradients and backpropagate them across multiple layers. Despite its success of existing work in accelerating propagation through sparseness, the relevant theoretical characteristics remain unexplored and we empirically find that they suffer from the loss of information contained in unpropagated gradients. To tackle these problems, in this work, we present a unified sparse backpropagation framework and provide a detailed analysis of its theoretical characteristics. Analysis reveals that when applied to a multilayer perceptron, our framework essentially performs gradient descent using an estimated gradient similar enough to the true gradient, resulting in convergence in probability under certain conditions. Furthermore, a simple yet effective algorithm named memorized sparse backpropagation (MSBP) is proposed to remedy the problem of information loss by storing unpropagated gradients in memory for the next learning. The experiments demonstrate that the proposed MSBP is able to effectively alleviate the information loss in traditional sparse backpropagation while achieving comparable acceleration.

This paper is concerned with the robustness of VAEs to adversarial attacks. We highlight that conventional VAEs are brittle under attack but that methods recently introduced for disentanglement such as $\beta$-TCVAE (Chen et al., 2018) improve robustness, as demonstrated through a variety of previously proposed adversarial attacks (Tabacof et al. (2016); Gondim-Ribeiro et al. (2018); Kos et al.(2018)). This motivated us to develop Seatbelt-VAE, a new hierarchical disentangled VAE that is designed to be significantly more robust to adversarial attacks than existing approaches, while retaining high quality reconstructions.

Motion planning is an essential component in most of today's robotic applications. In this work, we consider the learning setting, where a set of solved motion planning problems is used to improve the efficiency of motion planning on different, yet similar problems. This setting is important in applications with rapidly changing environments such as in e-commerce, among others. We investigate a general deep learning based approach, where a neural network is trained to map an image of the domain, the current robot state, and a goal robot state to the next robot state in the plan. We focus on the learning algorithm, and compare supervised learning methods with reinforcement learning (RL) algorithms. We first establish that supervised learning approaches are inferior in their accuracy due to insufficient data on the boundary of the obstacles, an issue that RL methods mitigate by actively exploring the domain. We then propose a modification of the popular DDPG RL algorithm that is tailored to motion planning domains, by exploiting the known model in the problem and the set of solved plans in the data. We show that our algorithm, dubbed DDPG-MP, significantly improves the accuracy of the learned motion planning policy. Finally, we show that given enough training data, our method can plan significantly faster on novel domains than off-the-shelf sampling based motion planners. Results of our experiments are shown in https://youtu.be/wHQ4Y4mBRb8.

Time-lagged autoencoders (TAEs) have been proposed as a deep learning regression-based approach to the discovery of slow modes in dynamical systems. However, a rigorous analysis of nonlinear TAEs remains lacking. In this work, we discuss the capabilities and limitations of TAEs through both theoretical and numerical analyses. Theoretically, we derive bounds for nonlinear TAE performance in slow mode discovery and show that in general TAEs learn a mixture of slow and maximum variance modes. Numerically, we illustrate cases where TAEs can and cannot correctly identify the leading slowest mode in two example systems: a 2D "Washington beltway" potential and the alanine dipeptide molecule in explicit water. We also compare the TAE results with those obtained using state-free reversible VAMPnets (SRVs) as a variational-based neural network approach for slow modes discovery, and show that SRVs can correctly discover slow modes where TAEs fail.