SAN FRANCISCO--We are already living with deep learning and large-scale neural networks, as evidenced by the growing number of applications that rely on computer vision, language understanding, and robotics. What we now want most from machine learning, said Google Senior Fellow Jeff Dean to the audience at SIGMOD 2016 keynote today (Tuesday, June 28), is --understanding." In a keynote talk, Dean outlined the history of machine learning (ML) and neural networks and various ways to program models to take advantage of raw data coming through in the form of images or audio. He also detailed how ML has taken shape at Google, which recently announced that it will open a machine learning center in Europe. The company developed its own accelerator chips for artificial intelligence it calls tensor processing units (TPUs) after the open source TensorFlow algorithms it released last year.

In this episode, Jennifer Marsman sits down and talks with Kushal Lakhotia and Walter Sun about the machine learning that powers Bing's predictions of who will win sporting matches like the NCAA March Madness tournament, the NBA playoffs, the NFL season, Wimbledon, the Cricket World Cup, the English Premier League, and more.

Ukip's only MP, Douglas Carswell, is the least friendly and trustworthy in the House of Commons, a social media survey has found. The newly appointed Shadow Health Secretary, Diane Abbott, was the most patronising, the same survey found. Artios, an artificial intelligence company, 'blind tested' 1,000 UK adults with social media content from an equal cross section of political parties. The highest rating for trustworthiness was only 14 per cent for a post by Prime Minister David Cameron. Lib Dem MP Greg Mudholland scored as the most friendly.

Prof Stephen Hawking, one of Britain's pre-eminent scientists, has said that efforts to create thinking machines pose a threat to our very existence. He told the BBC:"The development of full artificial intelligence could spell the end of the human race." His warning came in response to a question about a revamp of the technology he uses to communicate, which involves a basic form of AI. The theoretical physicist, who has the motor neurone disease amyotrophic lateral sclerosis (ALS), is using a new system developed by Intel to speak. Machine learning experts from the British company Swiftkey were also involved in its creation.

Fear not, Europhiles, for love is not lost. If you're downbeat about Britain's vote to leave the European Union, know that 48% of the country are with you, and one of them might just be your soulmate. No, honestly ... there is a new dating app in development that aims to let you meet other Remain voters so that you don't have to go through the pain alone. The idea for "Remainder" started off as a joke between "two ordinary voters" on Friday afternoon, but after receiving a huge number of sign-ups since the website went live, the pair are trying to make the app a reality. Billing itself as the "dating and social app for the 48%", Remainder is in the early-formation stage at the moment, but discussions with a dating app developer means a launch could be expected soon, the team behind it told the Guardian.

A matched formula is a CNF formula whose incidence graph admits a matching which matches a distinct variable to every clause. Such a formula is always satisfiable. Matched formulas are used, for example, in the area of parametrized complexity. We prove that the problem of counting the number of the models (satisfying assignments) of a matched formula is #P-complete. On the other hand, we define a class of formulas generalizing the matched formulas and prove that for a formula in this class one can choose in polynomial time a variable suitable for splitting the tree for the search of the models of the formula. As a consequence, the models of a formula from this class, in particular of any matched formula, can be generated sequentially with a delay polynomial in the size of the input. On the other hand, we prove that this task cannot be performed efficiently for linearly satisfiable formulas, which is a generalization of matched formulas containing the class considered above.

Recent work has shown how to improve delete relaxation heuristics by computing relaxed plans, i.e., the hFF heuristic, in a compiled planning task PiC which represents a given set C of fact conjunctions explicitly. While this compilation view of such partial delete relaxation is simple and elegant, its meaning with respect to the original planning task is opaque, and the size of PiC grows exponentially in |C|. We herein provide a direct characterization, without compilation, making explicit how the approach arises from a combination of the delete-relaxation with critical-path heuristics. Designing equations characterizing a novel view on h+ on the one hand, and a generalized version hC of hm on the other hand, we show that h+(PiC) can be characterized in terms of a combined hcplus equation. This naturally generalizes the standard delete-relaxation framework: understanding that framework as a relaxation over singleton facts as atomic subgoals, one can refine the relaxation by using the conjunctions C as atomic subgoals instead. Thanks to this explicit view, we identify the precise source of complexity in hFF(PiC), namely maximization of sets of supported atomic subgoals during relaxed plan extraction, which is easy for singleton-fact subgoals but is NP-complete in the general case. Approximating that problem greedily, we obtain a polynomial-time hCFF version of hFF(PiC), superseding the PiC compilation, and superseding the modified PiCce compilation which achieves the same complexity reduction but at an information loss. Experiments on IPC benchmarks show that these theoretical advantages can translate into empirical ones.

Two essential tasks in managing description logic knowledge bases are eliminating problematic axioms and incorporating newly formed ones. Such elimination and incorporation are formalised as the operations of contraction and revision in belief change. In this paper, we deal with contraction and revision for the DL-Lite family through a model-theoretic approach. Standard description logic semantics yields an infinite number of models for DL-Lite knowledge bases, thus it is difficult to develop algorithms for contraction and revision that involve DL models. The key to our approach is the introduction of an alternative semantics called type semantics which can replace the standard semantics in characterising the standard inference tasks of DL-Lite. Type semantics has several advantages over the standard one. It is more succinct and importantly, with a finite signature, the semantics always yields a finite number of models. We then define model-based contraction and revision functions for DL-Lite knowledge bases under type semantics and provide representation theorems for them. Finally, the finiteness and succinctness of type semantics allow us to develop tractable algorithms for instantiating the functions.

In the Compressed Sensing community, it is well known that given a matrix $X \in \mathbb R^{n\times p}$ with $\ell_2$ normalized columns, the Restricted Isometry Property (RIP) implies the Null Space Property (NSP). It is also well known that a small Coherence $\mu$ implies a weak RIP, i.e. the singular values of $X_T$ lie between $1-\delta$ and $1+\delta$ for "most" index subsets $T \subset \{1,\ldots,p\}$ with size governed by $\mu$ and $\delta$. In this short note, we show that a small Coherence implies a weak Null Space Property, i.e. $\Vert h_T\Vert_2 \le C \ \Vert h_{T^c}\Vert_1/\sqrt{s}$ for most $T \subset \{1,\ldots,p\}$ with cardinality $|T|\le s$. We moreover prove some singular value perturbation bounds that may also prove useful for other applications.

This paper proposes a variant of the method of Gu\'edon and Verhynin for estimating the cluster matrix in the Mixture of Gaussians framework via Semi-Definite Programming. A clustering oriented embedding is deduced from this estimate. The procedure is suitable for very high dimensional data because it is based on pairwise distances only. Theoretical garantees are provided and an eigenvalue optimisation approach is proposed for computing the embedding. The performance of the method is illustrated via Monte Carlo experiements and comparisons with other embeddings from the literature.