Machine learning models are often susceptible to adversarial perturbations of their inputs. Even small perturbations can cause state-of-the-art classifiers with high "standard" accuracy to produce an incorrect prediction with high confidence. To better understand this phenomenon, we study adversarially robust learning from the viewpoint of generalization. We show that already in a simple natural data model, the sample complexity of robust learning can be significantly larger than that of "standard" learning. This gap is information theoretic and holds irrespective of the training algorithm or the model family. We complement our theoretical results with experiments on popular image classification datasets and show that a similar gap exists here as well. We postulate that the difficulty of training robust classifiers stems, at least partially, from this inherently larger sample complexity.

Electricity is bought and sold in wholesale markets at prices that fluctuate significantly. Short-term forecasting of electricity prices is an important endeavor because it helps electric utilities control risk and because it influences competitive strategy for generators. As the "smart grid" grows, short-term price forecasts are becoming an important input to bidding and control algorithms for battery operators and demand response aggregators. While the statistics and machine learning literature offers many proposed methods for electricity price prediction, there is no consensus supporting a single best approach. We test two contrasting machine learning approaches for predicting electricity prices, regression decision trees and recurrent neural networks (RNNs), and compare them to a more traditional ARIMA implementation. We conduct the analysis on a challenging dataset of electricity prices from ERCOT, in Texas, where price fluctuation is especially high. We find that regression decision trees in particular achieves high performance compared to the other methods, suggesting that regression trees should be more carefully considered for electricity price forecasting.

Residential transformer population is a critical type of asset that many electric utility companies have been attempting to manage proactively and effectively to reduce unexpected failures and life losses that are often caused by transformer overloading. Within the typical power asset portfolio, the residential transformer asset is often large in population, having lowest reliability design, lacking transformer loading data and susceptible to customer loading behaviors such as adoption of distributed energy resources and electric vehicles. On the bright side, the availability of more residential operation data along with the advancement of data analytics techniques have provided a new path to further our understanding of local residential transformer overloading risks statistically. This research developed a new data-driven method to combine clustering analysis and the simulation of transformer temperature rise and insulation life loss to quantitatively and statistically assess the overloading risk of residential transformer population in one area and suggest proper risk management measures according to the assessment results. Case studies from an actual Canadian utility company have been presented and discussed in detail to demonstrate the applicability and usefulness of the proposed method.

In this contribution, we present a novel approach for segmenting laser radar (lidar) imagery into geometric time-height cloud locations with a fully convolutional network (FCN). We describe a semi-supervised learning method to train the FCN by: pre-training the classification layers of the FCN with "weakly labeled" lidar data, using "unsupervised" pre-training with the cloud locations of the Wang & Sassen (2001) cloud mask algorithm, and fully supervised learning with hand-labeled cloud locations. We show the model achieves higher levels of cloud identification compared to the cloud mask algorithm.

There are so many amazing ways artificial intelligence and machine learning are used behind the scenes to impact our everyday lives and inform business decisions and optimize operations for some of the world's leading companies. Here are 27 amazing practical examples of AI and machine learning. Using natural language processing, machine learning and advanced analytics, Hello Barbie listens and responds to a child. A microphone on Barbie's necklace records what is said and transmits it to the servers at ToyTalk. There, the recording is analyzed to determine the appropriate response from 8,000 lines of dialogue.

ABU DHABI – Japan and its top trade partner in the Middle East, the United Arab Emirates, agreed on Monday to expand economic, political and defense cooperation. Tokyo and Abu Dhabi also signed an investment protection agreement, capping off a two-day visit by Prime Minister Shinzo Abe to the oil-rich Gulf state. Abe arrived late Sunday on the first leg of a Middle East tour that will also take him to Jordan, Israel and the Palestinian territories. In a joint statement, the two countries praised growing trade between them. They "stressed the importance of further enhancing trade, investments, and business such as renewable energy, sustainable water desalination … artificial intelligence, health care and medical equipment," the statement said.

Global energy leader, BP is at the forefront of realizing the opportunities big data and artificial intelligence has for the energy industry. They use the technology to drive new levels of performance, improve the use of resources and safety and reliability of oil and gas production and refining. From sensors that relay the conditions at each site to using AI technology to improve operations, BP puts data at the fingertips of engineers, scientists and decision-makers to help drive high performance. In an attempt to deliver energy into the 21st century, GE Power uses big data, machine learning and Internet of Things (IoT) technology to build an "internet of energy." Advanced analytics and machine learning enable predictive maintenance and power, operations and business optimization to help GE Power work toward its vision of a "digital power plant."

Inspired by the matching of supply to demand in logistical problems, the optimal transport (or Monge--Kantorovich) problem involves the matching of probability distributions defined over a geometric domain such as a surface or manifold. In its most obvious discretization, optimal transport becomes a large-scale linear program, which typically is infeasible to solve efficiently on triangle meshes, graphs, point clouds, and other domains encountered in graphics and machine learning. Recent breakthroughs in numerical optimal transport, however, enable scalability to orders-of-magnitude larger problems, solvable in a fraction of a second. Here, we discuss advances in numerical optimal transport that leverage understanding of both discrete and smooth aspects of the problem. State-of-the-art techniques in discrete optimal transport combine insight from partial differential equations (PDE) with convex analysis to reformulate, discretize, and optimize transportation problems. The end result is a set of theoretically-justified models suitable for domains with thousands or millions of vertices. Since numerical optimal transport is a relatively new discipline, special emphasis is placed on identifying and explaining open problems in need of mathematical insight and additional research.

It has been an incredible few years watching artificial intelligence (AI) emerge from the world of academia to become a mainstream business practice, and even being mentioned in Teen Vogue. AI is arguably the biggest technology opportunity for the UK economy today, so it's good to see the government taking this seriously by making it a core part of its modern industrial strategy. Last Thursday, the government agreed its AI sector deal with international tech firms, which will see £1bn of investment put into the industry. The step-changes in the capacity to collect and process massive amounts of data, combined with unprecedented accessibility of scalable cloud computing power and a wave of global entrepreneurship, has unlocked the AI technology developed in the 1950s, leading to massive innovation. AI already enables me to plan traffic-free journeys, play the perfect next music track, unlock my iPhone using my face, set a timer using my voice – and soon, hopefully, drive better. It is being used to reduce energy consumption, to improve teaching in schools, and to help detect disease.

Machine Learning helps make complex systems more efficient. By applying advanced Machine Learning techniques such as Cognitive Fingerprinting, wind project operators can utilize these tools to learn from collected data, detect regular patterns, and optimize their own operations. In his session at 18th Cloud Expo, Stuart Gillen, Director of Business Development at SparkCognition, discussed how research has demonstrated the value of Machine Learning in delivering next generation analytics to improve safety, performance, and reliability in today's modern wind turbines. Speaker Bio Stuart Gillen is the Director of Business Development at SparkCognition. In this role, he is responsible for driving business engagements, partner development, marketing activities, and go-to market strategy.