Tag Archives: tell
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
More often than not, we fall into the trap of trying to predict and anticipate the future, forgetting that the future is up to us to envision and create. In the words of Buckminster Fuller, “We are called to be architects of the future, not its victims.”
But how, exactly, do we create a “good” future? What does such a future look like to begin with?
In Future Consciousness: The Path to Purposeful Evolution, Tom Lombardo analytically deconstructs how we can flourish in the flow of evolution and create a prosperous future for humanity. Scientifically informed, the books taps into themes that are constructive and profound, from both eastern and western philosophies.
As the executive director of the Center for Future Consciousness and an executive board member and fellow of the World Futures Studies Federation, Lombardo has dedicated his life and career to studying how we can create a “realistic, constructive, and ethical future.”
In a conversation with Singularity Hub, Lombardo discussed purposeful evolution, ethical use of technology, and the power of optimism.
Raya Bidshahri: Tell me more about the title of your book. What is future consciousness and what role does it play in what you call purposeful evolution?
Tom Lombardo: Humans have the unique capacity to purposefully evolve themselves because they possess future consciousness. Future consciousness contains all of the cognitive, motivational, and emotional aspects of the human mind that pertain to the future. It’s because we can imagine and think about the future that we can manipulate and direct our future evolution purposefully. Future consciousness empowers us to become self-responsible in our own evolutionary future. This is a jump in the process of evolution itself.
RB: In several places in the book, you discuss the importance of various eastern philosophies. What can we learn from the east that is often missing from western models?
TL: The key idea in the east that I have been intrigued by for decades is the Taoist Yin Yang, which is the idea that reality should be conceptualized as interdependent reciprocities.
In the west we think dualistically, or we attempt to think in terms of one end of the duality to the exclusion of the other, such as whole versus parts or consciousness versus physical matter. Yin Yang thinking is seeing how both sides of a “duality,” even though they appear to be opposites, are interdependent; you can’t have one without the other. You can’t have order without chaos, consciousness without the physical world, individuals without the whole, humanity without technology, and vice versa for all these complementary pairs.
RB: You talk about the importance of chaos and destruction in the trajectory of human progress. In your own words, “Creativity frequently involves destruction as a prelude to the emergence of some new reality.” Why is this an important principle for readers to keep in mind, especially in the context of today’s world?
TL: In order for there to be progress, there often has to be a disintegration of aspects of the old. Although progress and evolution involve a process of building up, growth isn’t entirely cumulative; it’s also transformative. Things fall apart and come back together again.
Throughout history, we have seen a transformation of what are the most dominant human professions or vocations. At some point, almost everybody worked in agriculture, but most of those agricultural activities were replaced by machines, and a lot of people moved over to industry. Now we’re seeing that jobs and functions are increasingly automated in industry, and humans are being pushed into vocations that involve higher cognitive and artistic skills, services, information technology, and so on.
RB: You raise valid concerns about the dark side of technological progress, especially when it’s combined with mass consumerism, materialism, and anti-intellectualism. How do we counter these destructive forces as we shape the future of humanity?
TL: We can counter such forces by always thoughtfully considering how our technologies are affecting the ongoing purposeful evolution of our conscious minds, bodies, and societies. We should ask ourselves what are the ethical values that are being served by the development of various technologies.
For example, we often hear the criticism that technologies that are driven by pure capitalism degrade human life and only benefit the few people who invented and market them. So we need to also think about what good these new technologies can serve. It’s what I mean when I talk about the “wise cyborg.” A wise cyborg is somebody who uses technology to serve wisdom, or values connected with wisdom.
RB: Creating an ideal future isn’t just about progress in technology, but also progress in morality. How we do decide what a “good” future is? What are some philosophical tools we can use to determine a code of ethics that is as objective as possible?
TL: Let’s keep in mind that ethics will always have some level of subjectivity. That being said, the way to determine a good future is to base it on the best theory of reality that we have, which is that we are evolutionary beings in an evolutionary universe and we are interdependent with everything else in that universe. Our ethics should acknowledge that we are fluid and interactive.
Hence, the “good” can’t be something static, and it can’t be something that pertains to me and not everybody else. It can’t be something that only applies to humans and ignores all other life on Earth, and it must be a mode of change rather than something stable.
RB: You present a consciousness-centered approach to creating a good future for humanity. What are some of the values we should develop in order to create a prosperous future?
TL: A sense of self-responsibility for the future is critical. This means realizing that the “good future” is something we have to take upon ourselves to create; we can’t let something or somebody else do that. We need to feel responsible both for our own futures and for the future around us.
Another one is going to be an informed and hopeful optimism about the future, because both optimism and pessimism have self-fulfilling prophecy effects. If you hope for the best, you are more likely to look deeply into your reality and increase the chance of it coming out that way. In fact, all of the positive emotions that have to do with future consciousness actually make people more intelligent and creative.
Some other important character virtues are discipline and tenacity, deep purpose, the love of learning and thinking, and creativity.
RB: Are you optimistic about the future? If so, what informs your optimism?
I justify my optimism the same way that I have seen Ray Kurzweil, Peter Diamandis, Kevin Kelly, and Steven Pinker justify theirs. If we look at the history of human civilization and even the history of nature, we see a progressive motion forward toward greater complexity and even greater intelligence. There’s lots of ups and downs, and catastrophes along the way, but the facts of nature and human history support the long-term expectation of continued evolution into the future.
You don’t have to be unrealistic to be optimistic. It’s also, psychologically, the more empowering position. That’s the position we should take if we want to maximize the chances of our individual or collective reality turning out better.
A lot of pessimists are pessimistic because they’re afraid of the future. There are lots of reasons to be afraid, but all in all, fear disempowers, whereas hope empowers.
Image Credit: Quick Shot / Shutterstock.com
We are a participant in the Amazon Services LLC Associates Program, an affiliate advertising program designed to provide a means for us to earn fees by linking to Amazon.com and affiliated sites. Continue reading