Compressing convolutional neural networks (CNNs) is essential for transferring the success of CNNs to a wide variety of applications to mobile devices. In contrast to directly recognizing subtle weights or filters as redundant in a given CNN, this paper presents an evolutionary method to automatically eliminate redundant convolution filters. We represent each compressed network as a binary individual of specific fitness. Then, the population is upgraded at each evolutionary iteration using genetic operations. As a result, an extremely compact CNN is generated using the fittest individual. In this approach, either large or small convolution filters can be redundant, and filters in the compressed network are more distinct. In addition, since the number of filters in each convolutional layer is reduced, the number of filter channels and the size of feature maps are also decreased, naturally improving both the compression and speed-up ratios. Experiments on benchmark deep CNN models suggest the superiority of the proposed algorithm over the state-of-the-art compression methods.

Deep models are the defacto standard in visual decision models due to their impressive performance on a wide array of visual tasks. However, they are frequently seen as opaque and are unable to explain their decisions. In contrast, humans can justify their decisions with natural language and point to the evidence in the visual world which led to their decisions. We postulate that deep models can do this as well and propose our Pointing and Justification (PJ-X) model which can justify its decision with a sentence and point to the evidence by introspecting its decision and explanation process using an attention mechanism. Unfortunately there is no dataset available with reference explanations for visual decision making. We thus collect two datasets in two domains where it is interesting and challenging to explain decisions. First, we extend the visual question answering task to not only provide an answer but also a natural language explanation for the answer. Second, we focus on explaining human activities which is traditionally more challenging than object classification. We extensively evaluate our PJ-X model, both on the justification and pointing tasks, by comparing it to prior models and ablations using both automatic and human evaluations.

Disney is experimenting with a deep learning AI that tracks movie goers' emotional reactions to films. The company's research branch developed neural networks that can assess viewers' reactions simply by monitoring their facial expressions as they watch movies like'Big Hero 6,' 'The Jungle Book' and'Star Wars: The Force Awakens.' While tests of the new method - called factorized variational autoencoders or FVAEs - are preliminary, it has already been demonstrated that that the new technique outperformed conventional methods. The data set ended up with 16 million facial landmarks from 3,179 viewers. Though preliminary, the experiment demonstrated a'very strong predictive performance' that could reliably guess a viewer's reactions to the remainder of a movie after just a few minutes Neural networks are computational models that learn similarly to humans, except with a lot more processing power.

If you've been keeping tabs on our recent AI coverage, you are probably aware that AI is the umbrella under which exists many concepts and methods of development. In the end, these different but not exclusive approaches seek to bestow machines with more human-like reasoning and intelligence. Deep Learning is one type of reasoning that AI platforms adopt. Deep Learning, as a representation of that concept, is receiving significant research efforts, investment, and even media buzz. Since the dawn of computing technology, developers created programs and algorithms by writing code that machines translate into precise instructions.

Rich Sutton is old school king of RL.

HoloLens contains a custom multiprocessor called the Holographic Processing Unit, or HPU. It is responsible for processing the information coming from all of the on-board sensors, including Microsoft's custom time-of-flight depth sensor, head-tracking cameras, the inertial measurement unit (IMU), and the infrared camera. The HPU is part of what makes HoloLens the world's first–and still only–fully self-contained holographic computer. Today, Harry Shum, executive vice president of our Artificial Intelligence and Research Group, announced in a keynote speech at CVPR 2017, that the second version of the HPU, currently under development, will incorporate an AI coprocessor to natively and flexibly implement DNNs. The chip supports a wide variety of layer types, fully programmable by us.

Artificial intelligence has been a growing field for tech companies and the technology is starting to make its way to consumer electronics. In a blog post Monday, Microsoft detailed plans for its HoloLens 2 mixed reality headset and confirmed that it would utilize a dedicated coprocessor for AI processing. Technologically, here's why Microsoft wants to move AI work right onto the HoloLens 2. Traditionally, companies have either offloaded AI processing onto the cloud or used the processor included in a device by default. But as AI applications have become increasingly demanding and resource-intensive, companies have run into the limitations of this approach. Especially at the consumer level, the processing time needed to bounce tasks to and from the cloud and the limits of existing hardware are hurdles to the lag-free AI performance that companies want from their devices.

As the ability and influence of Artificial Intelligence grows, so does its associated vocabulary. Due to AI's sheer complexity, it's becoming more and more important to understand the language used to describe it. Now, there are different types of AI itself. But what do developers and technologists really mean when they use these terms? It tells or programs computers to work in a certain way by writing logic into software.

Artificial Intelligence – Neil Jacobstein recently gave an information-packed talk at the Exponential Manufacturing conference on how artificial intelligence is redefining the future of work, production, supply chain, and design. Singularity University recently held the Exponential Manufacturing Summit with some of the world's brightest executives, entrepreneurs and investors being led through an intensive three-day program in Boston to prepare them for the changes brought forth by unstoppable technological progress. See the full lecture below. At the Summit, Neil Jacobstein chairs the Artificial Intelligence and Robotics Track at Singularity University, explored how exponential technologies including artificial intelligence, additive manufacturing, exponential energy, and bio manufacturing are continually redefining the future of work, production, supply chain, and design. "What you'll see when you look behind the scenes of most AI startups and even research labs is an emerging symbiosis between human intelligence and machine intelligence."

Deep learning is a crucial step toward achieving true AI--but the human brain still reigns supreme. This may be the first time in AI's history when a majority of experts agree the technology has practical value. From its conceptual beginnings in the 1950s led by legendary computer scientists such as Marvin Minsky and John McCarthy, AI's future viability has been the subject of fierce debate. As recently as 2000, the most proficient AI system was roughly comparable, in complexity, to the brain of a worm. Then, as high-bandwidth networking, cloud computing, and high-powered graphics-enabled microprocessors emerged, researchers began building multilayered neural networks--still extremely slow and limited compared to the human brain, but useful in practical ways.