We introduce DrummerNet, a drum transcription system that is trained in an unsupervised manner. DrummerNet does not require any ground-truth transcription and, with the data-scalability of deep neural networks, learns from a large unlabeled dataset. In DrummerNet, the target drum signal is first passed to a (trainable) transcriber, then reconstructed in a (fixed) synthesizer according to the transcription estimate. By training the system to minimize the distance between the input and the output audio signals, the transcriber learns to transcribe without ground truth transcription. Our experiment shows that DrummerNet performs favorably compared to many other recent drum transcription systems, both supervised and unsupervised.

Two of the most prominent public knowledge graphs, DBpedia [16] and YAGO [18], build rich taxonomies using Wikipedia's infoboxes and category graph, respectively. They describe more than five million entities and contain multiple hundred millions of triples [27]. When it comes to relation assertions (RAs), however, we observe - even for basic properties - a rather low coverage: More than 50% of the 1.35 million persons in DBpedia have no birthplace assigned; even more than 80% of birthplaces are missing in YAGO. At the same time, type assertions (TAs) are not present as well for many instances - for example, there are about half a million persons in DBpedia not explicitly typed as such [23]. Missing knowledge in Wikipedia-based knowledge graphs can be attributed to absent information in Wikipedia, but also to the extraction procedures of knowledge graphs. DBpedia uses infobox mappings to extract RAs for individual instances, but it does not explicate any information implicitly encoded in categories. YAGO uses manually defined patterns to assign RAs to entities of matching categories. For example, they extract a person's year of birth by

Artificial intelligence is all the buzz. A recent article by Jon Schuppe of NBC news asks if we are ready for police to use facial recognition software to track our every move. Well, ready or not, it's already starting to happen. The Maryland Image Repository System was used to identify the Annapolis Capital Gazette shooter. The shooter's image was captured on a security camera, and software was used to match that image to the repository, which includes driver's license photos as well as state and federal mug shots.

For decades the concept of artificial intelligence (AI) has required creative interpretations of scientific theory to capture people's attention. Whether it was the overwhelmingly capable character Data from "Star Trek," the humorously unnerved C-3PO from "Star Wars," or any other sci-fi AI manifestation, the fact still stands that intriguing forms of artificial intelligence have primarily been relegated to the entertainment world until only recently. However, the lack of capable forms of AI in everyday life has changed. With each passing year, increasingly complex artificial intelligence models are seeing the light of day. While they may not look like sophisticated androids at this point, many of these are the germs from which more advanced models will ultimately evolve.

One of the fundamental questions of cultural evolutionary research is how individual-level processes scale up to generate population-level patterns. Previous studies in music have revealed that frequency-based bias (e.g. conformity and novelty) drives large-scale cultural diversity in different ways across domains and levels of analysis. Music sampling is an ideal research model for this process because samples are known to be culturally transmitted between collaborating artists, and sampling events are reliably documented in online databases. The aim of the current study was to determine whether frequency-based bias has played a role in the cultural transmission of music sampling traditions, using a longitudinal dataset of sampling events across three decades. Firstly, we assessed whether turn-over rates of popular samples differ from those expected under neutral evolution. Next, we used agent-based simulations in an approximate Bayesian computation framework to infer what level of frequency-based bias likely generated the observed data. Despite anecdotal evidence of novelty bias, we found that sampling patterns at the population-level are most consistent with conformity bias.

We hear the term "digital transformation" all the time, and probably too often in my humble opinion. But what does it really mean and why does it appear to be a never-ending struggle that permeates our existence? Is it a task condemning us to repeat pushing a digital boulder up a mountain like Sisyphus, only to see it roll down and start all over again? Digital transformation is a term used by many and understood by few. So, in my usual manner (I am consistent if nothing else) I will also include another less used, but possible more descriptive term - digital disruption.

The expected reward in a linear stochastic bandit model is an unknown linear function of the chosen decision vector. In this paper, we consider the case where the expected reward is an unknown linear function of a projection of the decision vector onto a subspace. We call this the projection reward. Unlike the classical linear bandit problem, we assume that the projection reward is unobservable. Instead, the observed "reward" at each time step is the projection reward corrupted by another linear function of the decision vector projected onto a subspace orthogonal to the first. Such a model is useful in recommendation applications where the observed reward is corrupted by each individual's biases. In the case where there are finitely many decision vectors, we develop a strategy to achieve $O(\log T)$ regret, where $T$ is the number of time steps. In the case where the decision vector is chosen from an infinite compact set, our strategy achieves $O(T^{2/3}(\log{T})^{1/2})$ regret. Simulations verify the efficiency of our strategy.

Does this star have a planet? A new algorithm could help astronomers predict, on the basis of a star's chemical fingerprint, whether that star will host a giant gaseous exoplanet. "It's like Netflix," Natalie Hinkel, a planetary astrophysicist at the Southwest Research Institute in San Antonio, Texas, told Eos. Netflix "sees that you like goofy comedy, science fiction, and kung fu movies--a variety of different patterns" to predict whether you'll like a new movie. Likewise, her team's machine learning algorithm "will learn which elements are influential in deciding whether or not a star has a planet."

Newswire) VSBLTY Groupe Technologies Corp. (CSE: VSBY) (5VS.F) (VSBGF), a leading retail software and technology company, is teaming with Energetika, an international provider of "intelligent lighting" solutions, to install safety lighting and integrated security to Mexico City, and other Latin American cities designated as a "Smart City." Accessibility, habitability, sustainability, air quality, noise levels, energy, health and economic vitality are among the elements necessary to be selected as a "Smart City." Energetika is a leading provider of smart lighting solutions for economically efficient applications that incorporate security. Energetika chose VSBLTY to provide security technology that includes crowd analytics and facial recognition for residential, commercial and governmental applications. VSBLTY technology provides enhanced customer engagement and audience measurement using machine learning and computer vision.

This documentary projects the effects and dangers of Artificial Intelligence (AI) developments for next generations. The video addresses lots of examples in both negative and positive dimensions for using and developing Artificial Intelligence. In my opinion, one of the most important messages of this movie is that speakers in the movie believe that the development of AI is beneficial but it could misuse in lots of malicious areas. Example of that might be within war machines or development of mass destruction weapons which could seriously jeopardize our lives. The message of this movie is clearly states that machines can easily reproduce and duplicate themselves therefore development of full AI could spell the end of the human race.