Humans quickly solve tasks in novel systems with complex dynamics, without requiring much interaction. While deep reinforcement learning algorithms have achieved tremendous success in many complex tasks, these algorithms need a large number of samples to learn meaningful policies. In this paper, we present a task for navigating a marble to the center of a circular maze. While this system is very intuitive and easy for humans to solve, it can be very difficult and inefficient for standard reinforcement learning algorithms to learn meaningful policies. We present a model that learns to move a marble in the complex environment within minutes of interacting with the real system. Learning consists of initializing a physics engine with parameters estimated using data from the real system. The error in the physics engine is then corrected using Gaussian process regression, which is used to model the residual between real observations and physics engine simulations. The physics engine equipped with the residual model is then used to control the marble in the maze environment using a model-predictive feedback over a receding horizon. We contrast the learning behavior against the time taken by humans to solve the problem to show comparable behavior. To the best of our knowledge, this is the first time that a hybrid model consisting of a full physics engine along with a statistical function approximator has been used to control a complex physical system in real-time using nonlinear model-predictive control (NMPC). Codes for the simulation environment can be downloaded here https://www.merl.com/research/license/CME . A video describing our method could be found here https://youtu.be/xaxNCXBovpc .

The Arlo Pro 4 is a small but mighty outdoor home security camera. The Arlo Pro 4 Spotlight camera has higher video quality and better field of view than almost any camera we've tested--including the popular Nest Cam Outdoor. Other Arlo Pro 4 features include color night vision output, two-way talk capabilities, timely smart alerts, and easy integration with Amazon Alexa and Google Assistant. The Pro 4 is entirely wire-free and runs on a rechargeable battery that can last up to six months per charge. It also has a built-in spotlight that illuminates when motion is detected, and a smart siren that can be triggered automatically or remotely via the Arlo app.

Abstract: Rich functionalities of quantum and strongly correlated materials emerge from the interplay between the electronic, orbital, lattice, and spin degrees of freedom that often lead to complex structural and electronic phenomena spanning atomic to mesoscopic scales. In many cases, these phenomena are associated with translational symmetry breaking, local frozen disorder, or strongly correlated disorder. However, the relevant mechanisms and roles of individual subsystems often remain unknown. Over the last decade, Scanning Transmission Electron Microscopy has emerged as a powerful quantitative probe of materials structure on the atomic level, providing high veracity information on local chemical bonding, composition, and symmetry breaking distortions. We aim to harness the power of machine learning methods to build a comprehensive picture of the chemistry and physics of quantum materials from these observations.

If movies are to be believed, artificial intelligence is a one-way ticket to a dystopian future, with films like The Terminator, Blade Runner and The Matrix pointing to a bleak future for humanity – but new Australian research suggests AI could actually play a key role in avoiding the climate crisis. Australian researchers have unveiled a new artificial intelligence platform that has the potential to accelerate the development of cheaper and higher performance next generation solar cells, with the ability to discover new materials that do not exist yet. Researchers from the ARC Centre of Excellence in Exciton Science in Melbourne, have successfully demonstrated a new type of machine learning model that is able to predict the energy conversion efficiency of new materials, including those used in next generation organic solar cells. The model, developed by researchers based at RMIT University and Monash University, allows scientists to model'virtual materials' that do not yet exist, allowing progress towards the development of cheaper and higher performance solar cells to be fast-tracked. According to new research published in the journal Computational Materials, the new artificial intelligence platform is significantly faster than other machine learning programs, and its source code has been released freely for use by other researchers. The researchers believe the new model could help speed up the development of cheap and efficient organic solar cells, seen as a potentially cheaper alternative to traditional silicon based solar cells, but which have yet to achieve large-scale commercial deployment.

It seems the USA is determined to attest its dominance in the field of artificial intelligence. In August 2020, the Trump administration announced that The National Science Foundation and the Department of Energy have allocated $1 billion for advanced research in AI and quantum information. The investment will lead to the foundation of 12 new AI institutes and quantum information science (QIS) research institutes. The funds will be directed toward AI Research Institutes under the supervision of NSF and QIS Research Centers led by DOE. The $1 billion will be allocated for a period of five years in order to achieve advancements in fields like machine learning, computer vision, and quantum computing. The USA is aware of AI's importance for the "21st-century American workforce" and national economic growth.

Artificial intelligence can provide value to natural gas utilities and customers, a National Association of Regulatory Utility Commissioners (NARUC) primer states. The primer from the US regulatory non-profit is aimed to improve awareness of artificial intelligence tools and practices, with a focus on the potential to enhance natural gas utility performance. It zeroes in on the three most common challenges being faced. These are ageing distribution infrastructure, excavator damage to underground infrastructure and customer participation in energy efficiency programmes. Regarding ageing infrastructure, artificial intelligence can assist in identifying and prioritising repair and replacement programmes.

Spiking Neural Networks (SNNs) contain more biology-realistic structures and biology-inspired learning principles compared with that in standard Artificial Neural Networks (ANNs). The dynamic neurons in SNNs are non-differential, containing decayed historical states and generating event-based spikes after their states reaching the firing threshold. This dynamic characteristic of SNNs made it hard to be directly trained with standard back propagation (BP) which is considered not biologically plausible. In this paper, a Biologically-plausible Reward Propagation (BRP) algorithm is proposed and applied on a SNN architecture with both spiking-convolution (with both 1D and 2D convolutional kernels) and full-connection layers. Different with standard BP that propagated the error signals from post to pre synaptic neurons layer by layer, the BRP propagated the target labels instead of target errors directly from the output layer to all of the pre hidden layers. This effort was more consistent with the top-down reward-guiding learning in cortical columns of the neocortex. Then synaptic modifications with only local gradient differences were induced with pseudo-BP that might also be replaced with Spike-Timing Dependent Plasticity (STDP). The performance of the proposed BRP-SNN was further verified on spatial (including MNIST and Cifar-10) and temporal (including TIDigits and DvsGesture) tasks. The experimental result showed that the BRP played roles on convergent learning of SNN, reached higher accuracy compared with other state-of-the-art SNN algorithms, and saved more than 50% computational cost compared with that on ANNs. We think the introduction of biologically-plausible learning rules to the training procedure of biologically-realistic SNNs might give us more hints and inspirations towards a better understanding of the intelligent nature of the biological system.

Progressive digitalization is changing the game of many industrial sectors. Focus-ing on product quality the main profitability driver of this so-called Industry 4.0 will be the horizontal integration of information over the complete supply chain. Therefore, the European RFCS project 'Quality4.0' aims in developing an adap-tive platform, which releases decisions on product quality and provides tailored information of high reliability that can be individually exchanged with customers. In this context Machine Learning will be used to detect outliers in the quality data. This paper discusses the intermediate project results and the concepts developed so far for this horizontal integration of quality information.

Roof falls due to geological conditions are major safety hazards in mining and tunneling industries, causing lost work times, injuries, and fatalities. Several large-opening limestone mines in the Eastern and Midwestern United States have roof fall problems caused by high horizontal stresses. The typical hazard management approach for this type of roof fall hazard relies heavily on visual inspections and expert knowledge. In this study, we propose an artificial intelligence (AI) based system for the detection roof fall hazards caused by high horizontal stresses. We use images depicting hazardous and non-hazardous roof conditions to develop a convolutional neural network for autonomous detection of hazardous roof conditions. To compensate for limited input data, we utilize a transfer learning approach. In transfer learning, an already-trained network is used as a starting point for classification in a similar domain. Results confirm that this approach works well for classifying roof conditions as hazardous or safe, achieving a statistical accuracy of 86%. However, accuracy alone is not enough to ensure a reliable hazard management system. System constraints and reliability are improved when the features being used by the network are understood. Therefore, we used a deep learning interpretation technique called integrated gradients to identify the important geologic features in each image for prediction. The analysis of integrated gradients shows that the system mimics expert judgment on roof fall hazard detection. The system developed in this paper demonstrates the potential of deep learning in geological hazard management to complement human experts, and likely to become an essential part of autonomous tunneling operations in those cases where hazard identification heavily depends on expert knowledge.

We model the behavioral biases of human decision-making in securing interdependent systems and show that such behavioral decision-making leads to a suboptimal pattern of resource allocation compared to non-behavioral (rational) decision-making. We provide empirical evidence for the existence of such behavioral bias model through a controlled subject study with 145 participants. We then propose three learning techniques for enhancing decision-making in multi-round setups. We illustrate the benefits of our decision-making model through multiple interdependent real-world systems and quantify the level of gain compared to the case in which the defenders are behavioral. We also show the benefit of our learning techniques against different attack models. We identify the effects of different system parameters on the degree of suboptimality of security outcomes due to behavioral decision-making.