I suppose that whole idea about "make America great again" was only going to last just so long. It's been several weeks and we're a nation with a critically short attention span so it's probably time to start working on a new catchphrase. How about, "make human beings great again?" That seems to be what tech genius Elon Musk has in mind with his latest brainstorm. Rather than working on super fast trains or putting people on Mars, the entrepreneur has realized that the one thing truly holding the human race back is… human beings.

The future of the driverless car is much closer than people realize, Elon Musk reportedly said in a speech in Dubai Monday. The bad news, he points out, is that there will be a steep price to pay for the "great convenience." The Tesla TSLA, 3.94% and SpaceX CEO, according to CNBC, said the disruption will take place within about 20 years and will ultimately leave up to 15% of the global population without jobs. Musk also touched on one of his favorite topics: artificial intelligence. Specifically, he spoke of humans eventually merging with machines, like something out of a "Terminator" movie.

Like Google, Baidu's core service is also search – Baidu is said to account for 75% of search traffic in its homeland. Here, it has rolled out machine learning algorithms for voice and image recognition, as well as natural language processing, to help it return smarter, more useful and more personalized results. Baidu also makes its technology available to third parties such as other companies which want to benefit from the AI revolution but don't have the resources to develop their own algorithms and applications. Much of its software and systems have been made open source and it also provides access to its technology on an "as-a-service" basis. Businesses and organizations can use Baidu's systems to host their own data and run their own analytics projects in the cloud, paying only for the storage and computing resources which they use.

Machine intelligence is everywhere in facial recognition at airports to emotional sensing algorithms; machine generated Art work; legal and medical advisory search to sometimes fowl mouthed social chat bots. The Google company AI team recently announced they developed Google Neural Machine Translation system, GNMT, using a new technique that is improving results to near human translation speed accuracy. These advances that Google describe as machine translation at production scale, are testament to the rapid real-time advancement of AI into human experience and intelligence as well as beyond human capabilities. Andrew Ng of Stanford and Chief Scientist at Baidu Research famously said that word translation of 95% is 1 in every 20 words would likely be wrong, going to 99% is game changing. Andrew was quoted in a recent HBR article saying, "If a typical person can do a mental task with less than one second of thought, we can probably automate it using AI either now or in the near future."

We are in an age in which machine learning has increasing importance in our daily lives. Machine learning is put into action whenever your mobile map application automatically reminds you to leave for your next appointment because of unusual traffic situations. Besides personal assistants on your cell phones, wearable sport devices use machine-learning algorithms to propose personal training plans, and banks depend on accurate machine-learning models to detect malicious transactions. Healthcare, for instance, has also started to find helpful patterns in medical data using machine learning. Modern technologies allow for close monitoring of a patient's condition through a large volume of data provided by a number of sensors.

Coming into the new year, we've seen and heard plenty of buzz around AI. But the underlying data is often overlooked; it's a key foundational piece that impacts AI at scale. Therefore, the proverbial data statement, 'garbage in, garbage out,' and the implications of bad data quality, is arguably the most understated AI trend. "Successful models depend heavily on rich, deep client data. Many B2B marketers put themselves at a disadvantage by trying to make predictive work on a limited set of data or data that is just plain wrong."

Thank you for subscribing to the monthly THINK Marketing newsletter. Plenty of food for thought on the future of retail at NRF 2017, so much so that in the face of all the technology on show, Retail Reflections Founder and IBM Futurist, Andrew Busby, poses the fundamental question: Why retail? It seems everywhere we turn, we see the twin forces of consumer expectation and unrelenting pace and influence of technology are having a profound effect upon traditional retail business models. In the last 100 years, retail has hardly changed. If Harry Selfridge walked into his eponymous store in London today, he would, I'm sure, recognise the department store he opened in 1909.

TheTake, a site which launched as a way for consumers to buy that thing they saw in that movie, is set to begin selling an automated version of its service directly to businesses. The New York-based company is pitching studios and entertainment sites on a machine learning system that can identify products and locations as a way to generate revenue from product placements and experiential travel based on set locations. The new product is based on a year's worth of work that TheTake's development did to train a proprietary machine learning algorithm to identify images using a different technique than the industry standard, according to TheTake's chief executive Ty Cooper. Initially, the team behind TheTake would manually enter all the datasets and use an off-the-shelf computer visualization tool to identify images that fit the pre-defined parameters set by the company's staff. Companies like Universal Pictures, Comcast, Bravo, E!, Fandango, Sony Pictures and the Hallmark Channel, are testing out the AI-based service now, according to an email from Cooper.

If we have an initial learning rate high enough and a suitable decay factor for exponentially decaying the learning rate over a certain number of epoch, is it still need for us to manually explore the learning rate? Because if all goes well I believe the learning rate can automatically be sampled over a huge range of epoch. However, if we start off with a less than optimal learning rate, assuming the loss does not diverge to infinity, would the loss be less optimal than we have started with the optimal learning rate, even if we could reach the optimal learning rate through decaying the initial learning rate over time? Does the answer differ for a convex/non-convex loss? Specifically for deep learning problems, is an exponential decaying learning rate able to sample the learning rate better than done manually?

You may have noticed that I'm writing a little less about artificial intelligence than I had been. It may be that Skynet has imprisoned the real David Raab to block him from issuing dire warnings about its imminent threat to humanity and replaced him with a less alarmist simulation. You can't actually prove that isn't happening. But the David Raab, or Raab-bot, writing this will tell you it's because he's concluded that AI is destined to become so pervasive that it doesn't make sense to treat it as a distinct topic. It will simply be embedded in everything and so should be evaluated as part of whatever it belongs to. Lead Genius is a good example.