We introduce the forward-backward (FB) representation of the dynamics of a reward-free Markov decision process. It provides explicit near-optimal policies for any reward specified a posteriori. During an unsupervised phase, we use reward-free interactions with the environment to learn two representations via off-the-shelf deep learning methods and temporal difference (TD) learning. In the test phase, a reward representation is estimated either from reward observations or an explicit reward description (e.g., a target state). The optimal policy for thatreward is directly obtained from these representations, with no planning.

The leading approach to the simplex method, a widely used technique for balancing complex logistical constraints, can't get any better. In 1939, upon arriving late to his statistics course at UC Berkeley, George Dantzig--a first-year graduate student--copied two problems off the blackboard, thinking they were a homework assignment. He found the homework "harder to do than usual," he would later recount, and apologized to the professor for taking some extra days to complete it. A few weeks later, his professor told him that he had solved two famous open problems in statistics. Dantzig's work would provide the basis for his doctoral dissertation and, decades later, inspiration for the film .

This paper introduces a new meta-learning algorithm that combines population-based and point-based optimization. While population based approaches have been very popular in very rugged landscapes, current meta-learning methods are point-based and thus not suitable for optimizing such functions. This work presents two contributions, (1) a new architecture for population based meta-learning. This architecture, while more complicated, can be summarized as follows: each particle is composed of a set of 4 features (gradient, momentum, velocity, and attractions), an attention mechanism is applied to those features together with the hidden state. The outputs of the attention mechanism for all particles are fed into an inter-particle attention together with a similarity matrix.

Artificial intelligence (AI) is changing the face of business. No longer a futuristic concept, its impact is real. From tech giants like Google, Apple and Amazon to user-centric behemoths like Uber and Starbucks, everyone seems to be using AI technology to transform the customer experience (CX). But, it's not just corporate giants that are deploying AI. Smaller organizations are following suit.

This is a boon for retailers seeking to accurately predict demand, anticipate customer behavior, and optimize and personalize customer experiences. And as data volumes grow and processing power improves, machine learning becomes increasingly applicable in a wider range of retail areas to further optimize business processes and drive more impactful personalized and contextual consumer experiences and products. But McKinsey has assessed that the U.S. retail sector has only realized 30-40% of the potential margin improvements and productivity growth their analysts envisioned in 2011--and a large share of the value of this growth has gone to consumers through lower prices. So how will AI and machine learning change retail analytics, as they are currently defined?

Researchers from MIT's Computer Science and Artificial Intelligence Laboratory (CSAIL) and Columbia University are trying to make the process faster and easier: In a new paper, they've developed InstantCAD, a tool that lets designers interactively edit, improve, and optimize CAD models using a more streamlined and intuitive workflow. Traditional CAD systems are "parametric," which means that when engineers design models, they can change properties like shape and size ("parameters") based on different priorities. Matusik says InstantCAD could be particularly helpful for more intricate designs for objects like cars, planes, and robots, particularly for industries like car manufacturing that care a lot about squeezing every little bit of performance out of a product. "In a world where 3-D printing and industrial robotics are making manufacturing more accessible, we need systems that make the actual design process more accessible, too," Schulz says.

Artificial intelligence (AI) might soon become one of the biggest competitive differentiators for these businesses. Within the oil and gas industry, AI and machine learning are already being used for processing high volume data and to achieve operational efficiency, said Arunkumar Ranganathan, associate vice president and head of the domain and process consulting groups for energy, utilities, and services at technology consulting firm Infosys. "AI techniques are yet to be applied [for] interpreting geophysical and geological functions and in other core business functions," Ranganathan said. For example, AI is being used to optimize the drilling process and improve operational efficiency, leading to a reduction in drilling costs.

SQL is referred to as a declarative language as opposed to an imperative language like the 3GL's. You give it a high-level goal, and it figures out which machine learning algorithm to use, and tunes the hyperparameters for you. Are there other declarative machine learning systems out there? Their purpose is to allow non-Spark developers to write machine learning programs in languages they are comfortable in (like Python), yet be able to compile down to Spark Scala when the time comes to deploy to production.