NVIDIA debuted its Drive PX2 in-car supercomputer at CES in January, and now the company is showing off the Parker system on a chip powering it. The 256-core processor boasts up to 1.5 teraflops of juice for "deep learning-based self-driving AI cockpit systems," according to a post on NVIDIA's blog. That's in addition to 24 trillion deep learning operations per second it can churn out, too. For a perhaps more familiar touchpoint, NVIDIA says that Parker can also decode and encode 4K video streams running at 60FPS -- no easy feat on its own. However, Parker is significantly less beefy than NVIDIA's other deep learning initiative, the DGX-1 for Elon Musk's OpenAI, which can hit 170 teraflops of performance.

Neuroscience has focused on the detailed implementation of computation, studying neural codes, dynamics and circuits. In machine learning, however, artificial neural networks tend to eschew precisely designed codes, dynamics or circuits in favor of brute force optimization of a cost function, often using simple and relatively uniform initial architectures. Two recent developments have emerged within machine learning that create an opportunity to connect these seemingly divergent perspectives. First, structured architectures are used, including dedicated systems for attention, recursion and various forms of short- and long-term memory storage. Second, cost functions and training procedures have become more complex and are varied across layers and over time.

Intel Corp. INTC 0.34 % signaled it wants a bigger role in artificial intelligence, revealing plans to modify a line of chips to target a fast-growing market turning into a battleground for technology suppliers. The company told technology developers Wednesday that it plans next year to deliver a new version of the Xeon Phi processor--a product line previously targeted at scientific applications--with added features designed to accelerate tasks associated with what Silicon Valley calls artificial intelligence. Intel said the technology will help accelerate a technique called deep learning, increasingly used for tasks such as interpreting speech, identifying objects in photos and piloting autonomous vehicles. Intel's Xeon processors already are a fixture in data centers, and have a role in nearly all deep-learning tasks carried out there. But some users also install auxiliary processors for artificial-intelligence tasks, notably chips called GPUs that rival Nvidia Corp. NVDA 0.42 % has long sold for videogames.

Over the past few weeks, we've started to learn more and more about machine learning and the role it plays in computer vision, image classification, and deep learning. We've seen how Convolutional Neural Networks (CNNs) such as LetNet can be used to classify handwritten digits from the MNIST dataset. We've applied the k-NN algorithm to classify whether or not an image contains a dog or a cat. And we've learned how to apply hyperparameter tuning to optimize our model to obtain higher classification accuracy. However, there is another very important machine learning algorithm we have yet to explore -- one that can be built upon and extended naturally to Neural Networks and Convolutional Neural Networks.

My intern at TCL is over soon. Before going back to the campus for graduation, I have decided to build myself a personal deep learning rig. I guess I cannot really rely on the machines either in the company or in the lab, because ultimately the workstation is not mine, and the development environment may be messed up (It already happened once) . With a personal rig, I can conveniently use teamviewer to login my deep learning workstation at any time. And I got the chance to build everything from scratch.

Every summer, technologists turn to Gartner's Hype Cycle for Emerging Technologies, which has become a barometer of sorts for gauging the state of various hardware and software innovations that are expected to impact business and society over the next decade. This year, Gartner analysts have their eye on all manner of artificial intelligence technologies, including "smart machines" that can learn by themselves. After relieving "big data" from its hype-cycle duties last year–ostensibly due to the all-encompassing pervasiveness of data in this pervasively digital age–Gartner analysts this year are talking up a swath of related "smart machine" technologies. Together, Gartner refers to these technologies as key enablers of "the perceptual smart machine age" that is currently unfolding. "Smart machine technologies," the analyst group says in a press release, "will be the most disruptive class of technologies over the next 10 years due to radical computational power, near-endless amounts of data, and unprecedented advances in deep neural networks that will allow organizations with smart machine technologies to harness data in order to adapt to new situations and solve problems that no one has encountered previously."

A couple of years ago, machine learning suddenly started appearing on the horizon in enterprise software. Systems of engagement ceased to be the new frontier of innovation as'systems of intelligence' rapidly became all the rage. Nowadays, no self-respecting enterprise software vendor can be seen to be without a strategy for applying artifical intelligence to their systems. These AI-enriched, cloud-based systems promise a higher level of automation and productivity by discovering patterns in past behavior and then acting on them when the same conditions reoccur. Benefits are promised across every kind of enterprise function -- whether it's raising the success rate of salespeople, improving collections, planning projects more efficiently or fixing defective equipment before it fails.

In Part I of this series, we introduced the theory and intuition behind the VAE, an exciting development in machine learning for combined generative modeling and inference--"machines that imagine and reason." To recap: VAEs put a probabilistic spin on the basic autoencoder paradigm--treating their inputs, hidden representations, and reconstructed outputs as probabilistic random variables within a directed graphical model. With this Bayesian perspective, the encoder becomes a variational inference network, mapping observed inputs to (approximate) posterior distributions over latent space, and the decoder becomes a generative network, capable of mapping arbitrary latent coordinates back to distributions over the original data space. The beauty of this setup is that we can take a principled Bayesian approach toward building systems with a rich internal "mental model" of the observed world, all by training a single, cleverly-designed deep neural network. These benefits derive from an enriched understanding of data as merely the tip of the iceberg--the observed result of an underlying causative probabilistic process.

The ancient Chinese game of Go has simple rules, yet is extremely sophisticated. With a large board and few restrictions, the game is said to be a googol (10 to the hundredth power) times more complex than chess. There are more possible positions in Go than there are atoms in the universe. The game, in which opponents take turns placing black or white stones on a board, is played largely through intuition. Players understand that moves made early in the game can shape the match dozens of plays later. Go's subtleties, patterns, and elegance have captivated players, scholars, and mathematicians for millennia.

From July 20th to July 28th 2016, I had the opportunity of attending the 6th Lisbon Machine Learning School. The Lisbon Machine Learning School (LxMLS) is an annual event that brings together researchers and graduate students in the fields of NLP and Computational Linguistics, computer scientists with an interest in statistics and ML, and industry practitioners with a desire for a more in-depth understanding. Participants had a chance to join workshops and labs, where they got hands-on experience with building and exploring state-of-the-art deep learning models, as well as to attend talks and speeches by prominent deep learning and NLP researchers from a variety of academic and industrial organisations. You can find the entire programme here. In this blog post, I am going to share some of the highlights, key insights and takeaways of the summer school.