Deep neural networks and Deep Learning are powerful and popular algorithms. And a lot of their success lays in the careful design of the neural network architecture. I wanted to revisit the history of neural network design in the last few years and in the context of Deep Learning. For a more in-depth analysis and comparison of all the networks reported here, please see our recent article. Reporting top-1 one-crop accuracy versus amount of operations required for a single forward pass in multiple popular neural network architectures. It is the year 1994, and this is one of the very first convolutional neural networks, and what propelled the field of Deep Learning.

A future with highways full of self-driving cars or robot friends that can actually hold a decent conversation may not be far away. That's because we're living in the middle of an "artificial intelligence boom" -- a time when machines are becoming more and more like the human brain. That's partly because of an emerging subcategory of AI called "deep learning." It's a process that's often trying to mimic the human brain's neocortex, which helps humans with language processing, sensory perception and other functions. Essentially, deep learning is when machines figure out how to recognize objects.

In a recent post about Deep Learning applications, platforms, limitations and quantum computing we have announced a panel discussion on "How Artificial Intelligence is disrupting Industries" which took place on 25 May 2017 in Cape Town. This event was hosted by Far Ventures, a start-up studio and incubator that aims to foster technology entrepreneurship in Africa by founding and growing start-ups that can positively impact people's lives through technology. It was also supported by the Machine Intelligence Institute of Africa (MIIA), an innovative community and accelerator for Machine Intelligence and Data Science Research and Applications to help transform Africa. This post includes the links to the introductory presentation as well as the video of the panel discussion. Photos of the MIIA event can be found here. Photos of the MIIA event can be found here.

After winning its three-game match against Chinese grandmaster Ke Jie, the world's top Go player, AlphaGo is retiring. Demis Hassabis, the CEO and founder of DeepMind, the Google artificial intelligence lab that built this historic machine, tells WIRED he will now move the machine's designers to other projects. "This is some of the top people in the company," Hassabis says. "The idea is to really explore what we can do in other domains." Considering the world-shaking success of AlphaGo, that is a very powerful idea.

Jane Wang started out as an applied physicist modeling the complex network dynamics of memory systems in the brain before moving into experimental cognitive neuroscience as a postdoc at Northwestern. Since joining DeepMind two years ago, her non-machine learning background has equipped her with a unique set of tools and perspectives for tackling the hardest AI problems. "It's exhilarating to formulate theories of human brain function as powerful deep reinforcement learning models that can solve similarly complex tasks," she shares. Though Wang has been successful without a formal AI background, she's concerned the steep learning curve and hypercompetitive atmosphere of AI research can discourage diverse participation. "Although competitiveness drives the field forward, it also discourages those who wish to work in more inclusive, cooperative environments," she warns.

Think about the time it takes to see a doctor: Find a doctor, schedule an appointment, drive to the office, inevitably linger in the waiting room and then, finally, see the doctor. Artificial intelligence (AI)-powered wellness tools are reducing the steps required to receive a diagnosis. A wellness tool that uses image recognition and deep learning could enable you to self-diagnose a skin complaint. Leading technology and service providers (TSPs) are planning strategies based on an AI-enabled future. Apple, Dell, Facebook, Google, Microsoft and Samsung are some of the companies that have acquired AI startups and are developing new platforms for AI progress.

Protein-protein interactions (PPIs) are critical for many biological processes. It is therefore important to develop accurate high-throughput methods for identifying PPI to better understand protein function, disease occurrence, and therapy design. Though various computational methods for predicting PPI have been developed, their robustness for prediction with external datasets is unknown. Deep-learning algorithms have achieved successful results in diverse areas, but their effectiveness for PPI prediction has not been tested. We used a stacked autoencoder, a type of deep-learning algorithm, to study the sequence-based PPI prediction. The best model achieved an average accuracy of 97.19% with 10-fold cross-validation.

Time Budget: 30 seconds Hi, my name is Matthias Langer. I am currently a PhD student at La Trobe University. Today I would like to present to you Inferno, which is a deep learning system that we develop here in Melbourne and can run on top of Spark. Time Budget: 30 seconds My talk will be structured as follows: I will talk with you a little bit about DL. … then about DL and Spark… … our own DL system …. Time Budget: 30 seconds Talking Points: So without further ado, let's start… Time Budget: 1 minute So, what is deep learning? Deep learning is machine learning algorithm that tries to extract hierarchical features from input data. In itself that is kind of similar to how the brain does it in this slide. So how does that work: Let's say a stimulus (or input) comes from the eye and eventually ends up in region V1. There primitive features like edges are extracted.

The third reason is not that important performance-wise, but it does offer an additional insight into GPUs' undeniable supremacy over CPUs. The first part of the process involves fetching memory from the main or RAM memory and transferring it over to on-chip memory, or the L1 cache (instruction memory) and registers. Registers are attached directly to the execution unit, which for GPUs is the stream processor and for CPUs the core. This is where all the computation happens. Normally, you'd want both L1 and register memory to be as close to the execution engine and allow for a quick access by keeping the memories small.

Deep learning applies layers of hierarchical hidden variables to capture these interactions and nonlinearities. The theoretical roots lie in the Kolmogorov-Arnold representation theorem (Arnold, 1957; Kolmogorov, 1957) of multivariate functions, which states that any continuous multivariate function can be expressed as a superposition of continuous univariate semi-affine functions. This remarkable result has direct consequences for statistical modeling as a nonparametric pattern matching algorithm. Deep learning relies on pattern matching via its layers of univariate semi-affine functions and can be applied to both regression and classification problems. Deep learners provide a nonlinear predictor in complex settings where the input space can be very high dimensional.