5 Common Misconceptions about AI

In recent years I have ran into a number of misconceptions regarding AI, and sometimes when discussing AI with people from outside the field, I feel like we are talking about two different topics. This article is an attempt at clarifying what AI practitioners mean by AI, and where it is in its current state.
The first misconception has to do with Artificial General Intelligence, or AGI:

  1. Applied AI systems are just limited versions of AGI

Despite what many think,the state of the art in AI is still far behind human intelligence. Artificial General Intelligence, i.e. AGI, has been the motivating fuel for all AI scientists from Turing to today. Somewhat analogous to Alchemy, the eternal quest for AGI that replicates and exceeds human intelligence has resulted in the creation of many techniques and scientific breakthroughs. AGI has helped us understand facets of human and natural intelligence, and as a result, we’ve built  effective algorithms inspired by our understanding and models of them.
However, when it comes to practical applications of AI, AI practitioners do not necessarily restrict themselves to pure models of human decision making, learning, and problem solving. Rather, in the interest of solving the problem and achieving acceptable performance, AI practitioners often do what it takes to build practical systems. At the heart of the algorithmic breakthroughs that resulted in Deep Learning systems, for instance, is a technique called back-propagation. This technique, however, is not how the brain builds models of the world. This brings us to the next misconception:

  1. There is a one-size-fits-all AI solution.

A common misconception is that AI can be used to solve every problem out there–i.e. the state of the art AI has reached a level such that minor configurations of ‘the AI’ allows us to tackle different problems. I’ve even heard people assume that moving from one problem to the next makes the AI system smarter, as if the same AI system is now solving both problems at the same time. The reality is much different: AI systems need to be engineered, sometimes heavily,  and require specifically trained models in order to be applied to a problem. And while similar tasks, especially those involving sensing the world (e.g., speech recognition, image or video processing) now have a library of available reference models, these models need to be specifically engineered to meet deployment requirements and may not be useful out of the box. Furthermore, AI systems are seldom the only component of AI-based solutions. It often takes many tailor-made classically programed components to come together to augment one or more AI techniques used within a system. And yes, there are a multitude of different AI techniques out there, used alone or in hybrid solutions in conjunction with others, therefore it is incorrect to say:

  1. AI is the same as Deep Learning

Back in the day, we thought the term artificial neural networks (ANNs) was really cool. Until, that is, the initial euphoria around it’s potential backfired due to its lack of scaling and aptitude towards over-fitting. Now that those problems have, for the most part, been resolved, we’ve avoided the stigma of the old name by “rebranding” artificial neural networks as  “Deep Learning”. Deep Learning or Deep Networks are ANNs at scale, and the ‘deep’ refers not to deep thinking, but to the number of hidden layers we can now afford within our ANNs (previously it was a handful at most, and now they can be in the hundreds). Deep Learning is used to generate models off of labeled data sets. The ‘learning’ in Deep Learning methods refers to the generation of the models, not to the models being able to learn real-time as new data becomes available. The ‘learning’ phase of Deep Learning models actually happens offline, needs many iterations, is time and process intensive, and is difficult to parallelize.
Recently, Deep Learning models are being used in online learning applications. The online learning in such systems is achieved using different AI techniques such as Reinforcement Learning, or online Neuro-evolution. A limitation of such systems is the fact that the contribution from the Deep Learning model can only be achieved if the domain of use can be mostly experienced during the off-line learning period. Once the model is generated, it remains static and not entirely robust to changes in the application domain. A good example of this is in ecommerce applications–seasonal changes or short sales periods on ecommerce websites would require a deep learning model to be taken offline and retrained on sale items or new stock. However, now with platforms like Sentient Ascend that use evolutionary algorithms to power website optimization, large amounts of historical data is no longer needed to be effective, rather, it uses neuro-evolution to shift and adjust the website in real time based on the site’s current environment.   
For the most part, though, Deep Learning systems are fueled by large data sets, and so the prospect of new and useful models being generated from large and unique datasets has fueled the misconception that…

  1. It’s all about BIG data

It’s not. It’s actually about good data. Large, imbalanced datasets can be deceptive, especially if they only partially capture the data most relevant to the domain. Furthermore, in many domains, historical data can become irrelevant quickly. In high-frequency trading in the New York Stock Exchange, for instance, recent data is of much more relevance and value than, for example data from before 2001, when they had not yet adopted decimalization.
Finally, a general misconception I run into quite often:

  1. If a system solves a problem that we think requires intelligence, that means it is using AI

This one is a bit philosophical in nature, and it does depend on your definition of intelligence. Indeed, Turing’s definition would not refute this. However, as far as mainstream AI is concerned, a fully engineered system, say to enable self-driving cars, which does not use any AI techniques, is not considered an AI system. If the behavior of the system is not the result of the emergent behavior of AI techniques used under the hood, if programmers write the code from start to finish, in a deterministic and engineered fashion, then the system is not considered an AI-based system, even if it seems so.
AI paves the way for a better future
Despite the common misconceptions around AI, the one correct assumption is that AI is here to stay and is indeed, the window to the future. AI still has a long way to go before it can be used to solve every problem out there and to be industrialized for wide scale use. Deep Learning models, for instance, take many expert PhD-hours to design effectively, often requiring elaborately engineered parameter settings and architectural choices depending on the use case. Currently, AI scientists are hard at work on simplifying this task and are even using other AI techniques such as reinforcement learning and population-based or evolutionary architecture search to reduce this effort. The next big step for AI is to make it be creative and adaptive, while at the same time, powerful enough to exceed human capacity to build models.  
by Babak Hodjat, co-founder & CEO Sentient Technologies

What’s a Store For?

The first e-commerce transaction—a music CD, pizza, or weed, depending on who you ask—took place around thirty years ago. That means that first truly native ecommerce generation is now in charge of their own foot traffic and armed with at least one device that spares them the trouble of leaving the house. This, paired with the broader shift in consumer behavior across all generations, means brick and mortars need to find new ways to compete with digital to inspire visits and sales. Stores are evolving and, along the way, challenging the very notion of what a store is for.
Up against digital
A big part of brick and mortar’s evolution is digital integration. Today, retailers are working to enhance and personalize customer experience by connecting to consumers in-store through their mobile devices—building apps, targeting ads, and using beacons. You can find many examples of digital integration today, though online retailer Rebecca Minkoff’s flagship store in New York offers one of the more comprehensive ones; its interactive wall and dressing rooms have been credited with tripling expected clothing sales. Timberland also just launched its first connected store while Nordstrom’s commitment to digital integration has been credited with 50% growth in revenue over 5 years. (They just hired a former Amazon exec to serve as CTO.) Target, too, is getting into the mix, launching an LA25 initiative where it’s testing 50 of its top enhancements in 25 Los Angeles stores.
The IRL advantage
But digital integration is not the only strategy; retailers can also draw on the in-real-life [IRL] advantages of the physical space. Immediacy comes in here, with more retailers enabling online ordering and pick up in store or curbside. It’s competitive because fewer exclusively online retailers can offer this instant gratification, but is not necessarily a long-term strategy given that online fulfillment will continue to evolve and speed up.
More effective is the opportunity to build community. Oftentimes, this comes in the form of caffeine; Barnes and Noble was an early innovator here, adding a Starbucks to a New Jersey store back in 1993. Since then, many retailers have adopted or tested in-store cafes, including Urban Outfitters, Target, Restoration Hardware, and Kohl’s. Along the same lines, Target, Whole Foods, and Nordstrom, among others, are offering cocktails in some stores. When trying to attract customers and increase dwell time, there’s an advantage in offering something that can’t be instantly downloaded, like coffee, booze, and yes, maybe even tattoos. (See Whole Foods.)
Meanwhile, another concept that keeps popping up is—ahem—the pop up shop. The pop up shop’s currency is urgency; if customers don’t come now they risk missing out forever. Bloomingdales is hosting a pop up inspired by the musical Hamilton while Macy’s is bringing in pop ups as part of the reinvention of its Brooklyn store. The pop up also presents a low-risk testing ground for online retailers, one compelling example being Warby Parker’s touring store that was housed in a school bus.
But…is it a store?
As brick and mortar adapts, becoming deeper integrated with digital, acting a fulfillment center and expanding to offer drinks and other services, the classic definition of “store” begins to fragment. Already, the “store” has lost its longstanding position as the finale of the customer purchase funnel; in no small part because that purchase funnel itself is an antiquated concept. Savvy retailers and brands in general now think of the consumer experience as an ongoing loop, with consumers moving from digital to physical and back until, eventually, there may be no clear delineation between the two. This emphasis on the overall experience changes the expectations of stores. It also opens opportunities for more types of brands to invest in physical locations.
For example, last year, there was an more than an hour wait at the Museum of Feelings in downtown New York City. The museum invited visitors to walk through a sensory presentation of each feeling: Optimism, Joy, Invigorated, Exhilarated and Calm, while its exterior changed color to reflect the social mood of New York. You might argue that this wasn’t actually a store, but then it wasn’t actually a museum either; The Museum of Feelings was a branded retail experience for Glade, generating buzz for an otherwise not-so-buzzed-about brand.
More recently, Samsung launched Samsung 837, a “first-its-kind cultural destination, digital playground and marketing center of excellence.” Samsung 837 serves as a showcase for innovation, offering what may be the first virtual reality experience for many visitors and providing Instagram-friendly experiences like the walk-through Social Media Gallery. But what’s unique about Samsung’s space is that there is nothing sold there. It’s an experience—an opportunity for Samsung to tell its story and give visitors a way to get excited about the brand they’ll buy in the future.
In cases like these, brick and mortars serve as a marketing vehicle—an opportunity for brands to curate their own presence for customers, just as social provided the format to operate as a media company. It’s a trend that makes Amazon’s decision to open its own brick and mortars seem strategic. But is the return there?
It always comes back to data
The ability to more accurately track consumer activity gives brick and mortars a host of insights. Not only can the more connected store know what was purchased, they can also see what products compelled the most research, price comparisons, or inspired trips to the fitting room. They can engage with in-store customers via social media as well as encourage and measure posts from their store and, increasingly, tap into emotional analytics. Further, more sophisticated attribution measurement is making it possible to determine what investments drove traffic to the store, even without purchase.
Though it would be inaccurate to suggest that traffic and sales aren’t still the key performance indicators for most stores, this broader set of data, if put to use, can help a retailer optimize beyond the limits of its four walls—especially critical at a time when stores are closing so rapidly that CNN wrote “Store Closings are the Hottest Trend in Retail.”
Where to go from here
Digital has an odd way of creating challenges and then presenting solutions for those challenges it creates. It offers a range of ways of to add genuine value, from brand awareness to interaction, coupled with pop-up flexibility. If retailers are savvier about embracing this value, they’ll stand a better chance of attracting customers. If not, they’re not only missing out on opportunities in the near term, they’re limiting their future prospects for growth—after all, isn’t it a waste to see a store as a fulfilment outlet?

Etsy acquires French handmade ecommerce company A Little Market

Online marketplace Etsy, which has made its name in selling handmade goods from independent vendors, announced on Monday that it acquired A Little Market — an ecommerce shop based in France that also focuses on handmade goods. The Wall Street Journal reported that the acquisition is the largest from Etsy to date, although the terms of the deal have not been disclosed. The company says that A Little Market will continue to function independently, providing domestic goods in France. The company has been on a bit of a buying spree, as it acquired technology ecommerce company Grand St. in April of this year.

Still struggling, Fab will cut 80-90 employees tomorrow

It seems that hard times are continuing for struggling e-commerce startup Fab, as a company spokesperson confirmed to Buzzfeed today that it will lay of 80 to 90 employees — roughly a third of its current staff — in meetings tomorrow. The news comes as the team has spent the last six months putting together a line of sofas, which launched on Tuesday, in an effort to pivot from flash deals. The past year hasn’t been the greatest for Fab: Once worth roughly $1 billion to investors, the company shed 200 employees over two different layoffs in 2013 and also lost cofounder Bradford Shellhammer.