In recent years, great progress has been made in a variety of application domains thanks to the development of increasingly deeper neural networks. Unfortunately, the huge number of units of these networks makes them expensive both computationally and memory-wise. To overcome this, exploiting the fact that deep networks are over-parametrized, several compression strategies have been proposed. These methods, however, typically start from a network that has been trained in a standard manner, without considering such a future compression. In this paper, we propose to explicitly account for compression in the training process. To this end, we introduce a regularizer that encourages the parameter matrix of each layer to have low rank during training. We show that accounting for compression during training allows us to learn much more compact, yet at least as effective, models than state-of-the-art compression techniques.

Nowadays, the number of layers and of neurons in each layer of a deep network are typically set manually. While very deep and wide networks have proven effective in general, they come at a high memory and computation cost, thus making them impractical for constrained platforms. These networks, however, are known to have many redundant parameters, and could thus, in principle, be replaced by more compact architectures. In this paper, we introduce an approach to automatically determining the number of neurons in each layer of a deep network during learning. To this end, we propose to make use of a group sparsity regularizer on the parameters of the network, where each group is defined to act on a single neuron. Starting from an overcomplete network, we show that our approach can reduce the number of parameters by up to 80\% while retaining or even improving the network accuracy.

We conducted a comparison between the adjoint method, the Green's function method and a classical Gaussian process on the ordinary differential equation model presented in section 4.1.

In every election, after the polls close and the votes are counted, there comes a time for reflection. Pundits appear on cable news to offer theories, columnists pen op-eds with warnings and advice for the winners and losers, and parties conduct postmortems. The 2020 U.S. presidential election in which Donald Trump lost to Joe Biden was no exception. For Caltech undergrad Sreemanti Dey, the election offered a chance to do her own sort of reflection. Dey, an undergrad majoring in computer science, has a particular interest in using computers to better understand politics.

PLDT led discussions on Artificial Intelligence (AI) in helping drive business process optimization for the benefit of enhancing economic activities, progres, and recovery in Cebu during the recent ICT/BPM Summit. Along with other technical experts in the ICT sector, PLDT Technology Strategy and Transformation Office Technical Manager Paul Edward Alvarez tackled the untapped potential of AI in helping companies drive forward in their digital transformation journeys. "Demand for Machine learning and Ai and data skills continue to increase globally. For companies that have adopted AI, there have been very visible positive effects both on the top line with new services and cost savings in operations," said Alvarez. PLDT is currently exploring Artificial Intelligence to benefit and enrich the over-all customer experience of subscribers.

Why this might be the best job at the Olympics: You have a very important role. As a field support robot, you are charged with retrieving hammers and javelins and other thrown objects during the field events. After a hammer thrower throws her hammer, you zoom out onto the hammer pitch, retrieve the spent hammer, and bring it back to the general vicinity of the hammer-throwing circle, where it will eventually be thrown again. You are an important part of the Circle of Life, Hammer Throwing Edition. You don't just limit yourself to hammers and javelins: You also help out during the rugby events.

Last month, the British television network Channel 4 broadcast an "alternative Christmas address" by Queen Elizabeth II, in which the 94-year-old monarch was shown cracking jokes and performing a dance popular on TikTok. Of course, it wasn't real: The video was produced as a warning about deepfakes--apparently real images or videos that show people doing or saying things they never did or said. If an image of a person can be found, new technologies using artificial intelligence and machine learning now make it possible to show that person doing almost anything at all. The dangers of the technology are clear: A high-school teacher could be shown in a compromising situation with a student, a neighbor could be depicted as a terrorist. Can deepfakes, as such, be prohibited under American law?

This paper presents the Designer Preference Model, a data-driven solution that pursues to learn from user generated data in a Quality-Diversity Mixed-Initiative Co-Creativity (QD MI-CC) tool, with the aims of modelling the user's design style to better assess the tool's procedurally generated content with respect to that user's preferences. Through this approach, we aim for increasing the user's agency over the generated content in a way that neither stalls the user-tool reciprocal stimuli loop nor fatigues the user with periodical suggestion handpicking. We describe the details of this novel solution, as well as its implementation in the MI-CC tool the Evolutionary Dungeon Designer. We present and discuss our findings out of the initial tests carried out, spotting the open challenges for this combined line of research that integrates MI-CC with Procedural Content Generation through Machine Learning.

An attorney's computer program offers to screen potential jurors based on their ethnicity, political views and occupation to find a jury most favorable to a defense lawyer's case. Momus Analytics, the company was founded by attorney Alex Alvarez, trawls potential jurors' social media accounts and uses the findings to predict whether or not they should be chosen. The program includes a racially-biased algorithm that suggests Asian, Central American, and South American people are more likely to be leaders - a quality the program appears to prize. People who described their race as'other' were found to be likely to be leaders. Alvarez, who worked with Texas-based software designer Frogslayer to develop the program, has a pending patent application for the program.

The nature of applications and the way humans interact with them has changed based on the technologies available and the philosophies of the era. There's been a shift from monolithic enterprise applications to SaaS applications and apps, both of which provide users with more choices than they've had before. Instead of adapting work styles to an enterprise application's rigid design all the time, users have been able to choose SaaS options and apps that enable them to do their jobs faster and more effectively. Enter microservices, and the decomposition of applications is even more granular. "I think the biggest thing for any type of decomposition at this level is the key criteria for how granular you go and what makes sense," said Jimmy Pham, principal director of cloud platforms and application modernization at global management and IT consulting firm Booz Allen Hamilton. "From a pure application architect perspective, I always work with a team to define a domain-driven design, down to the context of what makes sense to be encapsulated in an independent function."