Neural Information Processing Systems http://nips.cc/

But assigning responsibility is complicated. These three visualizations help explain why. Take advantage of epic savings on award-winning reporting, razor-sharp analysis, and expert insights on your favorite technology topics. Subscribe today to save 50% on an annual subscription, plus receive a free digital copy of our "Generative AI and the future of work" report. This could be the cultivated meat industry's future: as a luxury product for the few.

GRT planner works in two phases. Although it did not gain any prize, it gave us good prospects for the future. STRIPS planners did not take part. The competition results have shown that the performance of the domain-independent heuristic planners is strongly affected by the representation of the domains. All GRT-related stuff is available at www.csd.auth.

In my last piece, Banking in the Age of Millennials, I wrote about the rise of a digitally-native generation, representing an immense opportunity for financial institutions. I also mentioned how Millennials differ from older age groups: they maintain low trust in banks, have a mobile-first expectation for financial products, and are more willing-to-try new products and services that take aim at a broader set of their financial needs. These findings led to 3 recommendations for developing Millennial products: hyper-personalization, build to rebuild, and the pursuit of value-add integrations and partnerships. I originally framed these tactics as a means for veteran financial institutions to rethink their product development. However, I spent most of my time at the Tech Crunch Disrupt Conference in September searching for startups that were leveraging the recommended approach.

hakey the Robot, conceived fifty years ago, was a seminal contribution to AI. Shakey perceived its world, planned how to achieve a goal, and acted to carry out that plan. This was revolutionary. At the Twenty-Ninth AAAI Conference on Artificial Intelligence, attendees gathered to celebrate Shakey, and to gain insights into how the AI revolution moves ahead. The celebration included a panel, chaired by Benjamin Kuipers and featuring AI pioneers Ed Feigenbaum, Peter Hart, and Nils Nilsson. This article includes written versions of the contributions of those panelists.

In many tasks related to reasoning about consequences of a logical theory, it is desirable to decompose the theory into a number of components with weakly-related or independent signatures. This facilitates reasoning when signature of a query formula belongs to only one of the components. However, an initial theory may be subject to change due to execution of actions affecting features mentioned in the theory. Having once computed a decomposition of a theory, one would like to know whether a decomposition has to be computed again for the theory obtained from taking into account the changes resulting from execution of an action. In the paper, we address this problem in the scope of the situation calculus, where change of an initial theory is related to the well-studied notion of progression. Progression provides a form of forward reasoning; it relies on forgetting values of those features which are subject to change and computing new values for them. We prove new results about properties of decomposition components under forgetting and show when a decomposition can be preserved in progression of an initial theory.

Planning based on propositional SAT(isfiability) is a powerful approach to computing step-optimal plans given a parallel execution semantics. In this setting: (i) a solution plan must be minimal in the number of plan steps required, and (ii) non-conflicting actions can be executed instantaneously in parallel at a plan step. Underlying SAT-based approaches is the invocation of a decision procedure on a SAT encoding of a bounded version of the problem. A fundamental limitation of existing approaches is the size of these encodings. This problem stems from the use of a direct representation of actions — i.e. each action has a corresponding variable in the encoding. A longtime goal in planning has been to mitigate this limitation by developing a more compact split — also termed lifted — representation of actions in SAT encodings of parallel step-optimal problems. This paper describes such a representation. In particular, each action and each parallel execution of actions is represented uniquely as a conjunct of variables. Here, each variable is derived from action pre and post- conditions . Because multiple actions share conditions , our encoding of the planning constraints is factored and relatively compact. We find experimentally that our encoding yields a much more efficient and scalable planning procedure over the state-of-the-art in a large set of planning benchmarks.