To gauge the complexity of the juggling act utility firms must perform to stay in business, consider two statistics. In the four years to 2018, the number of Britons who switched their energy supplier almost doubled to 5.9 million; at the same time, the contribution of renewables to energy firms' output mix grew from about 13% to just shy of 20%. One represents a fundamental change in consumer expectations of the service they receive while the other highlights the political and environmental pressures being brought to bear on suppliers' operations. To the list of challenges that are adding to the pressures under which utilities operate, add the tightening – and disparate – the grip of regulators, the fracturing of transmission networks and the increasing influence of activist investors. Managing these changing times can be incredibly challenging for established utilities, especially at a time when technology is enabling venture-backed start-ups to move into niche segments of their operations.

A research collaboration has achieved exaflop performance on the Summit supercomputer with a deep learning application used to model subsurface flow in the study of nuclear waste remediation. Their achievement, which will be presented during the "Deep Learning on Supercomputers" workshop at SC19, demonstrates the promise of physics-informed generative adversarial networks (GANs) for analyzing complex, large-scale science problems. The concept of physics-informed GANs is to encode prior information from physics into the neural network. This allows you to go well beyond the training domain, which is very important in applications where the conditions can change. GANs have been applied to model human face appearance with remarkable accuracy.

Improved Exploration through Latent Trajectory Optimization in Deep Deterministic Policy Gradient Kevin Sebastian Luck 1, Mel V ecerik 2, Simon Stepputtis 1, Heni Ben Amor 1 and Jonathan Scholz 2 Abstract -- Model-free reinforcement learning algorithms such as Deep Deterministic Policy Gradient (DDPG) often require additional exploration strategies, especially if the actor is of deterministic nature. This work evaluates the use of model-based trajectory optimization methods used for exploration in Deep Deterministic Policy Gradient when trained on a latent image embedding. In addition, an extension of DDPG is derived using a value function as critic, making use of a learned deep dynamics model to compute the policy gradient. This approach leads to a symbiotic relationship between the deep reinforcement learning algorithm and the latent trajectory optimizer . The trajectory optimizer benefits from the critic learned by the RL algorithm and the latter from the enhanced exploration generated by the planner . The developed methods are evaluated on two continuous control tasks, one in simulation and one in the real world. In particular, a Baxter robot is trained to perform an insertion task, while only receiving sparse rewards and images as observations from the environment. I NTRODUCTION Reinforcement learning (RL) methods enabled the development of autonomous systems that can autonomously learn and master a task when provided with an objective function. RL has been successfully applied to a wide range of tasks including flying [24], [17], manipulation [26], [9], [12], [3], [1], locomotion [10], [13], and even autonomous driving [6], [7].

Generative adversarial networks (GANs) are initially proposed to generate images by learning from a large number of samples. Recently, GANs have been used to emulate complex physical systems such as turbulent flows. However, a critical question must be answered before GANs can be considered trusted emulators for physical systems: do GANs-generated samples conform to the various physical constraints? These include both deterministic constraints (e.g., conservation laws) and statistical constraints (e.g., energy spectrum in turbulent flows). The latter have been studied in a companion paper (Wu et al. 2019. In the present work, we enforce deterministic yet approximate constraints on GANs by incorporating them into the loss function of the generator. We evaluate the performance of physics-constrained GANs on two representative tasks with geometrical constraints (generating points on circles) and differential constraints (generating divergence-free flow velocity fields), respectively. In both cases, the constrained GANs produced samples that precisely conform to the underlying constraints, even though the constraints are only enforced approximately. More importantly, the imposed constraints significantly accelerate the convergence and improve the robustness in the training. These improvements are noteworthy, as the convergence and robustness are two well-known obstacles in the training of GANs. Keywords: Generative adversarial networks, physics constraints, physics-informed machine learning 1. Introduction Machine learning and particularly deep learning has achieved significant success in a wide range of commercial domain applications such as image recognition, audio recognition, and natural language processing [1-5]. Corresponding author Email address: hengxiao@vt.edu For example, machine learning methods such as random forests and neural networks have been used to provide closure models for turbulent flows [6-9] and multiphase flows [10, 11] and to compute rock permeability directly from CT scan images [12]. They have also been used to discover ordinary and partial differential equations (ODEs and PDEs) from data [13-16]. Finally, neural networks have been used to solve exactly specified PDEs [17-20] and partially known PDEs by incorporating available data [21-24]. The scientific applications reviewed above mostly involve supervised learning, which consists of three steps: (a) postulate a model that maps inputs (features) to outputs (labels), controlled by a set of adjustable model parameters; (b) learn the parameters from training data (labeled examples of input-output pairs); and (c) use the fitted model to predict the responses for new inputs that were not included in the training data.

--Energy disaggregation, also referred to as a Non-Intrusive Load Monitoring (NILM), is the task of using an aggregate energy signal, for example coming from a whole-home power monitor, to make inferences about the different individual loads of the system. In this paper, we present a novel approach based on the encoder-decoder deep learning framework with an attention mechanism for solving NILM. The attention mechanism is inspired by the temporal attention mechanism that has been recently applied to get state-of-the-art results in neural machine translation, text summarization and speech recognition. The experiments have been conducted on two publicly available datasets AMPds and UK-DALE in seen and unseen conditions. The results show that our proposed deep neural network outperforms the state-of-the-art Denoising Auto-Encoder (DAE) proposed initially by Kelly and Knottenbely (2015) and its extended and improved architecture by Bonfigli et al. (2018), in all the addressed experimental conditions. We also show that modeling attention translates into the ability to correctly detect the state change of each appliance, that is of extreme interest in the field of energy disaggregation. Non-Intrusive Load Monitoring (NILM) is the task of estimating the power demand of each appliance given aggregate power demand signal recorded by a single electric meter monitoring multiple appliances [1]. In the last years, the research on NILM has been particularly active in the field of machine learning.

These iterative optimization algorithms are computational expensive and difficult to converge. Unlike iterative optimization methods, supervised machine learning using two stage training-testing becomes a great advantage for fast real-time inference since the most expensive computations are performed during training. Deep Learning plays a very important role tackling these type of problems but requires large dataset to train the multi-layer model parameters of the network [1]. Here, we are interested in creating a molecular image dataset including shape images from real space and diffraction patterns from reciprocal space for machine learning practices. We call this dataset Molecular-MNIST because it consists 10 different size of molecules where each molecule has 2,000 structural variants - in an analogy of the famous 10-digit handwritten dataset MNIST [2]. 2. Molecular-MNIST Dataset 2.1.

We optimize a six degrees of freedom hovering policy using reinforcement meta-learning. The policy maps flash LIDAR measurements directly to on/off spacecraft body-frame thrust commands, allowing hovering at a fixed position and attitude in the asteroid body-fixed reference frame. Importantly, the policy does not require position and velocity estimates, and can operate in environments with unknown dynamics, and without an asteroid shape model or navigation aids. Indeed, during optimization the agent is confronted with a new randomly generated asteroid for each episode, insuring that it does not learn an asteroid's shape, texture, or environmental dynamics. This allows the deployed policy to generalize well to novel asteroid characteristics, which we demonstrate in our experiments. The hovering controller has the potential to simplify mission planning by allowing asteroid body-fixed hovering immediately upon the spacecraft's arrival to an asteroid. This in turn simplifies shape model generation and allows resource mapping via remote sensing immediately upon arrival at the target asteroid.

Achieving such a massive scale and performance required full stack optimization and multiple strategies to extract maximum parallelism,

When it comes to the relationship between business development and technological innovation, we can generally separate two schools of thought. There are those who believe that technological progress is what propels businesses forward. And on the other hand, there are those who are certain that business investments are what makes innovations like contemporary geospatial AI possible. As with most opposing opinions – the truth is somewhere in between. Or, rather, the relations between cutting-edge tech and emerging business sectors are a never-ending circle; with business financing the research and development that enables the appearance of new tech, which in turn leads to new business opportunities and sectors.

Dan Jacobson, a research and development staff member in the Biosciences Division at the US Department of Energy's (DOE's) Oak Ridge National Laboratory (ORNL), has a few ideas. For the past 5 years, Jacobson and his team have studied plants to understand the genetic variables and patterns that make them adaptable to changing environments and climates. As a computational biologist, Jacobson uses some of the world's most powerful supercomputers for his work--including the recently decommissioned Cray XK7 Titan and the world's most powerful and smartest supercomputer for open science, the IBM AC922 Summit supercomputer, both located at the Oak Ridge Leadership Computing Facility (OLCF), a DOE Office of Science User Facility at ORNL. Last year, Jacobson and his team won an Association for Computing Machinery Gordon Bell Prize after using a special computing technique known as "mixed precision" on Summit to become the first group to reach exascale speed--approximately a quintillion calculations per second. Jacobson's team is currently working on numerous projects that form an integrated roadmap for the future of AI in plant breeding and bioenergy.