The challenge asks to provide a prediction for (i) a destination and the (ii) arrival time of ships in a streaming-fashion using Geo-spatial data in the maritime context. Novel aspects of our approach include the use of ensemble learning based on Random Forest, Gradient Boosting Decision Trees (GBDT), XGBoost Trees and Extremely Randomized Trees (ERT) in order to provide a prediction for a destination while for the arrival time, we propose the use of Feed-forward Neural Networks. In our evaluation, we were able to achieve an accuracy of 97% for the port destination classification problem and 90% (in mins) for the ETA prediction.

Efficiently scheduling data processing jobs on distributed compute clusters requires complex algorithms. Current systems, however, use simple generalized heuristics and ignore workload structure, since developing and tuning a bespoke heuristic for each workload is infeasible. In this paper, we show that modern machine learning techniques can generate highly-efficient policies automatically. Decima uses reinforcement learning (RL) and neural networks to learn workload-specific scheduling algorithms without any human instruction beyond specifying a high-level objective such as minimizing average job completion time. Off-the-shelf RL techniques, however, cannot handle the complexity and scale of the scheduling problem. To build Decima, we had to develop new representations for jobs' dependency graphs, design scalable RL models, and invent new RL training methods for continuous job arrivals. Our prototype integration with Spark on a 25-node cluster shows that Decima outperforms several heuristics, including hand-tuned ones, by at least 21%. Further experiments with an industrial production workload trace demonstrate that Decima delivers up to a 17% reduction in average job completion time and scales to large clusters.

This paper presents a technology for simple and computationally efficient improvements of a generic Artificial Intelligence (AI) system, including Multilayer and Deep Learning neural networks. The improvements are, in essence, small network ensembles constructed on top of the existing AI architectures. Theoretical foundations of the technology are based on Stochastic Separation Theorems and the ideas of the concentration of measure. We show that, subject to mild technical assumptions on statistical properties of internal signals in the original AI system, the technology enables instantaneous and computationally efficient removal of spurious and systematic errors with probability close to one on the datasets which are exponentially large in dimension. The method is illustrated with numerical examples and a case study of ten digits recognition from American Sign Language.

Deep reinforcement learning (DRL) has achieved outstanding results in recent years. This has led to a dramatic increase in the number of applications and methods. Recent works have explored learning beyond single-agent scenarios and have considered multiagent scenarios. Initial results report successes in complex multiagent domains, although there are several challenges to be addressed. In this context, first, this article provides a clear overview of current multiagent deep reinforcement learning (MDRL) literature. Second, it provides guidelines to complement this emerging area by (i) showcasing examples on how methods and algorithms from DRL and multiagent learning (MAL) have helped solve problems in MDRL and (ii) providing general lessons learned from these works. We expect this article will help unify and motivate future research to take advantage of the abundant literature that exists in both areas (DRL and MAL) in a joint effort to promote fruitful research in the multiagent community.

Doctoral students at Stevens Institute of Technology hope the robot they're developing will be able to dive into waters and perform tasks that could be very dangerous for humans. "We would like the robot to be able to do infrastructure inspections, ideally to assess the integrity of underwater infrastructure, make sure everything is intact, working properly, that there are no damages or defects. Or potentially from a security standpoint, that there are no anomalies planted on an underwater piece of infrastructure," said Dr. Brendan Englot, professor of mechanical engineering. To do this, the robot must be able to understand its location in the water and be able to accurately find the structures it needs to assess. Through a process called machine learning, the robot gathers data and improves its own performance.

I don't know why the developers made that decision, although the map is certainly cluttered enough without these Historical Locations. It's too bad you could play through the whole of Assassin's Creed: Odyssey ($60 on Humble) without stumbling on this other layer though. There's a lot of interesting historical information contained within, and a lot of great pseudo-historical information as well--insight into the Ancient Greece of Odyssey that explains how and why it differs from the strictly historical Ancient Greece we might know.

Provost Ian Walmsley welcomed Education Secretary Damian Hinds to the College's South Kensington Campus The Department for Education's leadership team were given an insight into the College's pioneering research, education and innovation. On Thursday 11 October Provost Ian Walmsley and President Alice Gast welcomed the Secretary of State for Education, Damian Hinds MP, to the College for the Department for Education board's away day. Before the board meeting began, Anne Milton MP, Minister of State for Apprenticeships and Skills, a number of senior civil servants and non-executive directors from the Department for Education met Provost Ian Walmsley, Vice-Provost (Education) Professor Simone Buitendijk and a number of students and academics from across the College. The visit included tours and demonstrations at the Aerial Robotics Lab, the Carbon Capture Pilot Plant and the Dyson School of Design Engineering. Director Dr Mirko Kovac gave a presentation on the work of the Aerial Robotics Lab at Imperial's Department of Aeronautics.

Amazon has scrapped a "sexist" tool that used artificial intelligence to decide the best candidates to hire for jobs. Members of the team working on the system said it effectively taught itself that male candidates were preferable. The artificial intelligence software was created by a team at Amazon's Edinburgh office in 2014 as a way to automatically sort through CVs and select the most talented applicants. But the algorithm rapidly taught itself to favour male candidates over female ones, according to members of the team who spoke to Reuters. Amazon wage increase could result in lower pay for some employees Black Friday 2018: The best Amazon deals Will Amazon's deliver-on-demand smart homes be the future of housing? Will Amazon's deliver-on-demand smart homes be the future of housing?

In recognition of the fact that their obituary pages had been dominated by white men, in 2018 the New York Times published an obituary of the Countess Ada Lovelace. Alongside Grace Hopper and Katherine Johnson, Lovelace has become an icon for women in technology. So much so that the second Tuesday in October is recognised internationally as Ada Lovelace Day. Lovelace was from a wealthy background; her father was the poet Lord Byron and her mother, Anne Isabella Milbanke, the "princess of parallelograms", was a keen mathematician and social reformer. Social scientists of today would describe Lovelace as having high "science capital" – her well-connected parents meant her mentors and advisers were members of the British scientific elite, including the polymaths Mary Somerville and Charles Babbage.

Amazon worked on building an artificial-intelligence tool to help with hiring, but the plans backfired when the company discovered the system discriminated against women, Reuters reports. Citing five sources, Reuters said Amazon set up an engineering team in Edinburgh, Scotland, in 2014 to find a way to automate its recruitment. The company created 500 computer models to trawl through past candidates' résumés and pick up on about 50,000 key terms. The system would crawl the web to recommend candidates. "They literally wanted it to be an engine where I'm going to give you 100 résumés, it will spit out the top five, and we'll hire those," one source told Reuters.