Google's Project Soli plays a key role in the magic. Soli is designed for what the project calls "virtual tools," in which a gesture mimics the familiar interaction with a physical tool. In this case, of course, the gesture is the rubbing of thumb and forefinger associated with playing a tiny violin. Another key piece of the World's Tiniest Violin is an open-source machine-learning tool called the Wekinator.

In this post you will discover how you can check-point your deep learning models during training in Python using the Keras library. When training deep learning models, the checkpoint is the weights of the model. Checkpointing is setup to save the network weights only when there is an improvement in classification accuracy on the validation dataset (monitor'val_acc' and mode'max'). In this post you have discovered the importance of checkpointing deep learning models for long training runs.

Zuckerberg was taking questions live from Facebook users Tuesday as part of a demo of his company's live video feature, which lets anyone launch live global broadcasts using their web cameras. Seinfeld, being Seinfeld, appeared less interested in talking about connecting the world and new products and more interested in probingZuckerberg's personal life. The break is small and Zuckerberg, Facebook CEO, isn't wearing a cast. The comedian noted that Facebook staff had repeatedly warned him about the CEO's broken arm as he wandered around Facebook's Menlo Park, California, campus on Tuesday.

On Monday, Apple exec Craig Federighi quickly mentioned how Apple is going to try and catch up to Google's smart Photo app: "on-device intelligence." Similarly, Microsoft offers developers a cloud service that will automatically caption photos. "It's disappointing to see that Apple is not offering developers and customers the option to tap into the power of applying machine learning to data hosted in cloud services. Apple has increased the privacy and encryption capabilities in their operating systems, but at the expense of not taking advantage of powerful machine learning capabilities," he adds.

Well, if I have a look at my personal life, there are some work categories where AI will be of great value to me--a maid, a driver, an on call doctor, even an assistant to help me filter through the information I have to churn through. When it comes down to it, I don't think AI can replace human communication, a combination of emotional and analytical perspective. We know when we're interacting with a machine, and as complex as we can make AI, will we ever truly feel it is more than talking to a toaster? Camilla Urdahl works for unified communications company Unified Inbox, and will speak to strangers at events about communications strategies, new business models, and how best to accelerate innovation.

The tech giant decided not to source its chips from other providers, instead building its own customised processors, dubbed Tensor Processing Units (TPUs), for its machine learning software. "We've been running TPUs inside our data centers for more than a year, and have found them to deliver an order of magnitude better-optimized performance per watt for machine learning," Google said in a blog post. Since AI has only recently found commercial use cases, companies that process AI functions are still solely using graphics processing units (GPUs) to do the job. "TPU is tailored to machine learning applications, allowing the chip to be more tolerant of reduced computational precision, which means it requires fewer transistors per operation," Google said.

Clearly, the idea of Artificial Intelligence (AI) has been with us in pop culture for a while. One of their breakthroughs involves the use of a technique called deep learning. Though it takes many forms, one example of deep learning is the creation of electronic "neural networks" that can mimic different basic operations occurring in real webs of neurons. This advancement enables the machine to take direct inputs from the world without human involvement and create its own internal representation for further processing."

Norm Jouppi, a distinguished hardware engineer at Google, detailed the company's public disclosure of the Tensor Processing Unit (TPU) last week after the CEO Sundar Pichai's earlier announcement at Google I/O. Several questions came up around how the TensorFlow-optimized chipset could compete with publicly available hardware like Nvidia's Tesla P100 and even PaaS providers like Nervana that provide machine learning services. Google's public disclosure of the TPU may have been related to Nvidia's release of the Tesla P100 in April. Jouppi noted that Google wants to lead the industry in machine learning and make the innovation available to its customers, but didn't disclose specific plans or offerings to do so at this time.

Alexy Khabrov's talk at LX Scala caused the Scala community to raise their arms for a revolution. After his talk, Alexy discussed his thoughts on how to ease the transition for Python data scientists into the Scala community, what Scala can learn from Python as a programming language, and how machine learning will influence the future of the Scala community within the next five years.