Tag Archives: psychology
Dr. Been Kim wants to rip open the black box of deep learning.
A senior researcher at Google Brain, Kim specializes in a sort of AI psychology. Like cognitive psychologists before her, she develops various ways to probe the alien minds of artificial neural networks (ANNs), digging into their gory details to better understand the models and their responses to inputs.
The more interpretable ANNs are, the reasoning goes, the easier it is to reveal potential flaws in their reasoning. And if we understand when or why our systems choke, we’ll know when not to use them—a foundation for building responsible AI.
There are already several ways to tap into ANN reasoning, but Kim’s inspiration for unraveling the AI black box came from an entirely different field: cognitive psychology. The field aims to discover fundamental rules of how the human mind—essentially also a tantalizing black box—operates, Kim wrote with her colleagues.
In a new paper uploaded to the pre-publication server arXiv, the team described a way to essentially perform a human cognitive test on ANNs. The test probes how we automatically complete gaps in what we see, so that they form entire objects—for example, perceiving a circle from a bunch of loose dots arranged along a clock face. Psychologist dub this the “law of completion,” a highly influential idea that led to explanations of how our minds generalize data into concepts.
Because deep neural networks in machine vision loosely mimic the structure and connections of the visual cortex, the authors naturally asked: do ANNs also exhibit the law of completion? And what does that tell us about how an AI thinks?
Enter the Germans
The law of completion is part of a series of ideas from Gestalt psychology. Back in the 1920s, long before the advent of modern neuroscience, a group of German experimental psychologists asked: in this chaotic, flashy, unpredictable world, how do we piece together input in a way that leads to meaningful perceptions?
The result is a group of principles known together as the Gestalt effect: that the mind self-organizes to form a global whole. In the more famous words of Gestalt psychologist Kurt Koffka, our perception forms a whole that’s “something else than the sum of its parts.” Not greater than; just different.
Although the theory has its critics, subsequent studies in humans and animals suggest that the law of completion happens on both the cognitive and neuroanatomical level.
Take a look at the drawing below. You immediately “see” a shape that’s actually the negative: a triangle or a square (A and B). Or you further perceive a 3D ball (C), or a snake-like squiggle (D). Your mind fills in blank spots, so that the final perception is more than just the black shapes you’re explicitly given.
Image Credit: Wikimedia Commons contributors, the free media repository.
Neuroscientists now think that the effect comes from how our visual system processes information. Arranged in multiple layers and columns, lower-level neurons—those first to wrangle the data—tend to extract simpler features such as lines or angles. In Gestalt speak, they “see” the parts.
Then, layer by layer, perception becomes more abstract, until higher levels of the visual system directly interpret faces or objects—or things that don’t really exist. That is, the “whole” emerges.
The Experiment Setup
Inspired by these classical experiments, Kim and team developed a protocol to test the Gestalt effect on feed-forward ANNs: one simple, the other, dubbed the “Inception V3,” far more complex and widely used in the machine vision community.
The main idea is similar to the triangle drawings above. First, the team generated three datasets: one set shows complete, ordinary triangles. The second—the “Illusory” set, shows triangles with the edges removed but the corners intact. Thanks to the Gestalt effect, to us humans these generally still look like triangles. The third set also only shows incomplete triangle corners. But here, the corners are randomly rotated so that we can no longer imagine a line connecting them—hence, no more triangle.
To generate a dataset large enough to tease out small effects, the authors changed the background color, image rotation, and other aspects of the dataset. In all, they produced nearly 1,000 images to test their ANNs on.
“At a high level, we compare an ANN’s activation similarities between the three sets of stimuli,” the authors explained. The process is two steps: first, train the AI on complete triangles. Second, test them on the datasets. If the response is more similar between the illusory set and the complete triangle—rather than the randomly rotated set—it should suggest a sort of Gestalt closure effect in the network.
Right off the bat, the team got their answer: yes, ANNs do seem to exhibit the law of closure.
When trained on natural images, the networks better classified the illusory set as triangles than those with randomized connection weights or networks trained on white noise.
When the team dug into the “why,” things got more interesting. The ability to complete an image correlated with the network’s ability to generalize.
Humans subconsciously do this constantly: anything with a handle made out of ceramic, regardless of shape, could easily be a mug. ANNs still struggle to grasp common features—clues that immediately tells us “hey, that’s a mug!” But when they do, it sometimes allows the networks to better generalize.
“What we observe here is that a network that is able to generalize exhibits…more of the closure effect [emphasis theirs], hinting that the closure effect reflects something beyond simply learning features,” the team wrote.
What’s more, remarkably similar to the visual cortex, “higher” levels of the ANNs showed more of the closure effect than lower layers, and—perhaps unsurprisingly—the more layers a network had, the more it exhibited the closure effect.
As the networks learned, their ability to map out objects from fragments also improved. When the team messed around with the brightness and contrast of the images, the AI still learned to see the forest from the trees.
“Our findings suggest that neural networks trained with natural images do exhibit closure,” the team concluded.
That’s not to say that ANNs recapitulate the human brain. As Google’s Deep Dream, an effort to coax AIs into spilling what they’re perceiving, clearly demonstrates, machine vision sees some truly weird stuff.
In contrast, because they’re modeled after the human visual cortex, perhaps it’s not all that surprising that these networks also exhibit higher-level properties inherent to how we process information.
But to Kim and her colleagues, that’s exactly the point.
“The field of psychology has developed useful tools and insights to study human brains– tools that we may be able to borrow to analyze artificial neural networks,” they wrote.
By tweaking these tools to better analyze machine minds, the authors were able to gain insight on how similarly or differently they see the world from us. And that’s the crux: the point isn’t to say that ANNs perceive the world sort of, kind of, maybe similar to humans. It’s to tap into a wealth of cognitive psychology tools, established over decades using human minds, to probe that of ANNs.
“The work here is just one step along a much longer path,” the authors conclude.
“Understanding where humans and neural networks differ will be helpful for research on interpretability by enlightening the fundamental differences between the two interesting species.”
Image Credit: Popova Alena / Shutterstock.com Continue reading
Scarcely a day goes by without another headline about neural networks: some new task that deep learning algorithms can excel at, approaching or even surpassing human competence. As the application of this approach to computer vision has continued to improve, with algorithms capable of specialized recognition tasks like those found in medicine, the software is getting closer to widespread commercial use—for example, in self-driving cars. Our ability to recognize patterns is a huge part of human intelligence: if this can be done faster by machines, the consequences will be profound.
Yet, as ever with algorithms, there are deep concerns about their reliability, especially when we don’t know precisely how they work. State-of-the-art neural networks will confidently—and incorrectly—classify images that look like television static or abstract art as real-world objects like school-buses or armadillos. Specific algorithms could be targeted by “adversarial examples,” where adding an imperceptible amount of noise to an image can cause an algorithm to completely mistake one object for another. Machine learning experts enjoy constructing these images to trick advanced software, but if a self-driving car could be fooled by a few stickers, it might not be so fun for the passengers.
These difficulties are hard to smooth out in large part because we don’t have a great intuition for how these neural networks “see” and “recognize” objects. The main insight analyzing a trained network itself can give us is a series of statistical weights, associating certain groups of points with certain objects: this can be very difficult to interpret.
Now, new research from UCLA, published in the journal PLOS Computational Biology, is testing neural networks to understand the limits of their vision and the differences between computer vision and human vision. Nicholas Baker, Hongjing Lu, and Philip J. Kellman of UCLA, alongside Gennady Erlikhman of the University of Nevada, tested a deep convolutional neural network called VGG-19. This is state-of-the-art technology that is already outperforming humans on standardized tests like the ImageNet Large Scale Visual Recognition Challenge.
They found that, while humans tend to classify objects based on their overall (global) shape, deep neural networks are far more sensitive to the textures of objects, including local color gradients and the distribution of points on the object. This result helps explain why neural networks in image recognition make mistakes that no human ever would—and could allow for better designs in the future.
In the first experiment, a neural network was trained to sort images into 1 of 1,000 different categories. It was then presented with silhouettes of these images: all of the local information was lost, while only the outline of the object remained. Ordinarily, the trained neural net was capable of recognizing these objects, assigning more than 90% probability to the correct classification. Studying silhouettes, this dropped to 10%. While human observers could nearly always produce correct shape labels, the neural networks appeared almost insensitive to the overall shape of the images. On average, the correct object was ranked as the 209th most likely solution by the neural network, even though the overall shapes were an exact match.
A particularly striking example arose when they tried to get the neural networks to classify glass figurines of objects they could already recognize. While you or I might find it easy to identify a glass model of an otter or a polar bear, the neural network classified them as “oxygen mask” and “can opener” respectively. By presenting glass figurines, where the texture information that neural networks relied on for classifying objects is lost, the neural network was unable to recognize the objects by shape alone. The neural network was similarly hopeless at classifying objects based on drawings of their outline.
If you got one of these right, you’re better than state-of-the-art image recognition software. Image Credit: Nicholas Baker, Hongjing Lu, Gennady Erlikhman, Philip J. Kelman. “Deep convolutional networks do not classify based on global object shape.” Plos Computational Biology. 12/7/18. / CC BY 4.0
When the neural network was explicitly trained to recognize object silhouettes—given no information in the training data aside from the object outlines—the researchers found that slight distortions or “ripples” to the contour of the image were again enough to fool the AI, while humans paid them no mind.
The fact that neural networks seem to be insensitive to the overall shape of an object—relying instead on statistical similarities between local distributions of points—suggests a further experiment. What if you scrambled the images so that the overall shape was lost but local features were preserved? It turns out that the neural networks are far better and faster at recognizing scrambled versions of objects than outlines, even when humans struggle. Students could classify only 37% of the scrambled objects, while the neural network succeeded 83% of the time.
Humans vastly outperform machines at classifying object (a) as a bear, while the machine learning algorithm has few problems classifying the bear in figure (b). Image Credit: Nicholas Baker, Hongjing Lu, Gennady Erlikhman, Philip J. Kelman. “Deep convolutional networks do not classify based on global object shape.” Plos Computational Biology. 12/7/18. / CC BY 4.0
“This study shows these systems get the right answer in the images they were trained on without considering shape,” Kellman said. “For humans, overall shape is primary for object recognition, and identifying images by overall shape doesn’t seem to be in these deep learning systems at all.”
Naively, one might expect that—as the many layers of a neural network are modeled on connections between neurons in the brain and resemble the visual cortex specifically—the way computer vision operates must necessarily be similar to human vision. But this kind of research shows that, while the fundamental architecture might resemble that of the human brain, the resulting “mind” operates very differently.
Researchers can, increasingly, observe how the “neurons” in neural networks light up when exposed to stimuli and compare it to how biological systems respond to the same stimuli. Perhaps someday it might be possible to use these comparisons to understand how neural networks are “thinking” and how those responses differ from humans.
But, as yet, it takes a more experimental psychology to probe how neural networks and artificial intelligence algorithms perceive the world. The tests employed against the neural network are closer to how scientists might try to understand the senses of an animal or the developing brain of a young child rather than a piece of software.
By combining this experimental psychology with new neural network designs or error-correction techniques, it may be possible to make them even more reliable. Yet this research illustrates just how much we still don’t understand about the algorithms we’re creating and using: how they tick, how they make decisions, and how they’re different from us. As they play an ever-greater role in society, understanding the psychology of neural networks will be crucial if we want to use them wisely and effectively—and not end up missing the woods for the trees.
Image Credit: Irvan Pratama / Shutterstock.com Continue reading
The human mind can be a confusing and overwhelming place. Despite incredible leaps in human progress, many of us still struggle to make our peace with our thoughts. The roots of this are complex and multifaceted. To find explanations for the global mental health epidemic, one can tap into neuroscience, psychology, evolutionary biology, or simply observe the meaningless systems that dominate our modern-day world.
This is not only the context of our reality but also that of the critically-acclaimed Netflix series, Maniac. Psychological dark comedy meets science fiction, Maniac is a retro, futuristic, and hallucinatory trip that is filled with hidden symbols. Directed by Cary Joji Fukunaga, the series tells the story of two strangers who decide to participate in the final stage of a “groundbreaking” pharmaceutical trial—one that combines novel pharmaceuticals with artificial intelligence, and promises to make their emotional pain go away.
Naturally, things don’t go according to plan.
From exams used for testing defense mechanisms to techniques such as cognitive behavioral therapy, the narrative infuses genuine psychological science. As perplexing as the series may be to some viewers, many of the tools depicted actually have a strong grounding in current technological advancements.
Catalysts for Alleviating Suffering
In the therapy of Maniac, participants undergo a three-day trial wherein they ingest three pills and appear to connect their consciousness to a superintelligent AI. Each participant is hurled into the traumatic experiences imprinted in their subconscious and forced to cope with them in a series of hallucinatory and dream-like experiences.
Perhaps the most recognizable parallel that can be drawn is with the latest advancements in psychedelic therapy. Psychedelics are a class of drugs that alter the experience of consciousness, and often cause radical changes in perception and cognitive processes.
Through a process known as transient hypofrontality, the executive “over-thinking” parts of our brains get a rest, and deeper areas become more active. This experience, combined with the breakdown of the ego, is often correlated with feelings of timelessness, peacefulness, presence, unity, and above all, transcendence.
Despite being not addictive and extremely difficult to overdose on, regulators looked down on the use of psychedelics for decades and many continue to dismiss them as “party drugs.” But in the last few years, all of this began to change.
Earlier this summer, the FDA granted breakthrough therapy designation to MDMA for the treatment of PTSD, after several phases of successful trails. Similar research has discovered that Psilocybin (also known as magic mushrooms) combined with therapy is far more effective than traditional forms of treatment to treat depression and anxiety. Today, there is a growing and overwhelming body of research that proves that not only are psychedelics such as LSD, MDMA, or Psylicybin effective catalysts to alleviate suffering and enhance the human condition, but they are potentially the most effective tools out there.
It’s important to realize that these substances are not solutions on their own, but rather catalysts for more effective therapy. They can be groundbreaking, but only in the right context and setting.
In Maniac, the medication-assisted therapy is guided by what appears to be a super-intelligent form of artificial intelligence called the GRTA, nicknamed Gertie. Gertie, who is a “guide” in machine form, accesses the minds of the participants through what appears to be a futuristic brain-scanning technology and curates customized hallucinatory experiences with the goal of accelerating the healing process.
Such a powerful form of brain-scanning technology is not unheard of. Current levels of scanning technology are already allowing us to decipher dreams and connect three human brains, and are only growing exponentially. Though they are nowhere as advanced as Gertie (we have a long way to go before we get to this kind of general AI), we are also seeing early signs of AI therapy bots, chatbots that listen, think, and communicate with users like a therapist would.
The parallels between current advancements in mental health therapy and the methods in Maniac can be startling, and are a testament to how science fiction and the arts can be used to explore the existential implications of technology.
Not Necessarily a Dystopia
While there are many ingenious similarities between the technology in Maniac and the state of mental health therapy, it’s important to recognize the stark differences. Like many other blockbuster science fiction productions, Maniac tells a fundamentally dystopian tale.
The series tells the story of the 73rd iteration of a controversial drug trial, one that has experienced many failures and even led to various participants being braindead. The scientists appear to be evil, secretive, and driven by their own superficial agendas and deep unresolved emotional issues.
In contrast, clinicians and researchers are not only required to file an “investigational new drug application” with the FDA (and get approval) but also update the agency with safety and progress reports throughout the trial.
Furthermore, many of today’s researchers are driven by a strong desire to contribute to the well-being and progress of our species. Even more, the results of decades of research by organizations like MAPS have been exceptionally promising and aligned with positive values. While Maniac is entertaining and thought-provoking, viewers must not forget the positive potential of such advancements in mental health therapy.
Science, technology, and psychology aside, Maniac is a deep commentary on the human condition and the often disorienting states that pain us all. Within any human lifetime, suffering is inevitable. It is the disproportionate, debilitating, and unjust levels of suffering that we ought to tackle as a society. Ultimately, Maniac explores whether advancements in science and technology can help us live not a life devoid of suffering, but one where it is balanced with fulfillment.
Image Credit: xpixel / Shutterstock.com Continue reading
Exponential technologies (AI, VR, 3D printing, and networks) are radically reshaping traditional retail.
E-commerce giants (Amazon, Walmart, Alibaba) are digitizing the retail industry, riding the exponential growth of computation.
Many brick-and-mortar stores have already gone bankrupt, or migrated their operations online.
Massive change is occurring in this arena.
For those “real-life stores” that survive, an evolution is taking place from a product-centric mentality to an experience-based business model by leveraging AI, VR/AR, and 3D printing.
Let’s dive in.
Last year, 3.8 billion people were connected online. By 2024, thanks to 5G, stratospheric and space-based satellites, we will grow to 8 billion people online, each with megabit to gigabit connection speeds.
These 4.2 billion new digital consumers will begin buying things online, a potential bonanza for the e-commerce world.
At the same time, entrepreneurs seeking to service these four-billion-plus new consumers can now skip the costly steps of procuring retail space and hiring sales clerks.
Today, thanks to global connectivity, contract production, and turnkey pack-and-ship logistics, an entrepreneur can go from an idea to building and scaling a multimillion-dollar business from anywhere in the world in record time.
And while e-commerce sales have been exploding (growing from $34 billion in Q1 2009 to $115 billion in Q3 2017), e-commerce only accounted for about 10 percent of total retail sales in 2017.
In 2016, global online sales totaled $1.8 trillion. Remarkably, this $1.8 trillion was spent by only 1.5 billion people — a mere 20 percent of Earth’s global population that year.
There’s plenty more room for digital disruption.
AI and the Retail Experience
For the business owner, AI will demonetize e-commerce operations with automated customer service, ultra-accurate supply chain modeling, marketing content generation, and advertising.
In the case of customer service, imagine an AI that is trained by every customer interaction, learns how to answer any consumer question perfectly, and offers feedback to product designers and company owners as a result.
Facebook’s handover protocol allows live customer service representatives and language-learning bots to work within the same Facebook Messenger conversation.
Taking it one step further, imagine an AI that is empathic to a consumer’s frustration, that can take any amount of abuse and come back with a smile every time. As one example, meet Ava. “Ava is a virtual customer service agent, to bring a whole new level of personalization and brand experience to that customer experience on a day-to-day basis,” says Greg Cross, CEO of Ava’s creator, an Austrian company called Soul Machines.
Predictive modeling and machine learning are also optimizing product ordering and the supply chain process. For example, Skubana, a platform for online sellers, leverages data analytics to provide entrepreneurs constant product performance feedback and maintain optimal warehouse stock levels.
Blockchain is set to follow suit in the retail space. ShipChain and Ambrosus plan to introduce transparency and trust into shipping and production, further reducing costs for entrepreneurs and consumers.
Meanwhile, for consumers, personal shopping assistants are shifting the psychology of the standard shopping experience.
Amazon’s Alexa marks an important user interface moment in this regard.
Alexa is in her infancy with voice search and vocal controls for smart homes. Already, Amazon’s Alexa users, on average, spent more on Amazon.com when purchasing than standard Amazon Prime customers — $1,700 versus $1,400.
As I’ve discussed in previous posts, the future combination of virtual reality shopping, coupled with a personalized, AI-enabled fashion advisor will make finding, selecting, and ordering products fast and painless for consumers.
But let’s take it one step further.
Imagine a future in which your personal AI shopper knows your desires better than you do. Possible? I think so. After all, our future AIs will follow us, watch us, and observe our interactions — including how long we glance at objects, our facial expressions, and much more.
In this future, shopping might be as easy as saying, “Buy me a new outfit for Saturday night’s dinner party,” followed by a surprise-and-delight moment in which the outfit that arrives is perfect.
In this future world of AI-enabled shopping, one of the most disruptive implications is that advertising is now dead.
In a world where an AI is buying my stuff, and I’m no longer in the decision loop, why would a big brand ever waste money on a Super Bowl advertisement?
The dematerialization, demonetization, and democratization of personalized shopping has only just begun.
The In-Store Experience: Experiential Retailing
In 2017, over 6,700 brick-and-mortar retail stores closed their doors, surpassing the former record year for store closures set in 2008 during the financial crisis. Regardless, business is still booming.
As shoppers seek the convenience of online shopping, brick-and-mortar stores are tapping into the power of the experience economy.
Rather than focusing on the practicality of the products they buy, consumers are instead seeking out the experience of going shopping.
The Internet of Things, artificial intelligence, and computation are exponentially improving the in-person consumer experience.
As AI dominates curated online shopping, AI and data analytics tools are also empowering real-life store owners to optimize staffing, marketing strategies, customer relationship management, and inventory logistics.
In the short term,retail store locations will serve as the next big user interface for production 3D printing (custom 3D printed clothes at the Ministry of Supply), virtual and augmented reality (DIY skills clinics), and the Internet of Things (checkout-less shopping).
In the long term,we’ll see how our desire for enhanced productivity and seamless consumption balances with our preference for enjoyable real-life consumer experiences — all of which will be driven by exponential technologies.
One thing is certain: the nominal shopping experience is on the verge of a major transformation.
The convergence of exponential technologies has already revamped how and where we shop, how we use our time, and how much we pay.
Twenty years ago, Amazon showed us how the web could offer each of us the long tail of available reading material, and since then, the world of e-commerce has exploded.
And yet we still haven’t experienced the cost savings coming our way from drone delivery, the Internet of Things, tokenized ecosystems, the impact of truly powerful AI, or even the other major applications for 3D printing and AR/VR.
Perhaps nothing will be more transformed than today’s $20 trillion retail sector.
Hold on, stay tuned, and get your AI-enabled cryptocurrency ready.
Abundance Digital Online Community: I’ve created a digital/online community of bold, abundance-minded entrepreneurs called Abundance Digital.
Abundance Digital is my ‘onramp’ for exponential entrepreneurs — those who want to get involved and play at a higher level. Click here to learn more.
Image Credit: Zapp2Photo / Shutterstock.com Continue reading