Phenomenal essay by Lucas Kunce, a U.S. Marine who served in Iraq and Afghanistan, responding to news that 4,600 Google employees signed a petition urging the company to refuse to build weapons technology:
People frequently threw objects of all sizes at our vehicles in anger and protest. Aside from roadside bombs, the biggest threat at the time, particularly in crowded areas, was an armor-piercing hand-held grenade. It looked like a dark soda can with a handle protruding from the bottom. Or, from a distance and with only an instant to decide, it looked just like many of the other objects that were thrown at us.
One day in Falluja, at the site of a previous attack, an Iraqi man threw a dark oblong object at one of the vehicles in my sister team. The Marine in the turret, believing it was an armor-piercing grenade, shot the man in the chest. The object turned out to be a shoe.
[. . . . .]
When I think about A.I. and weapons development, I don’t imagine Skynet, the Terminator, or some other Hollywood dream of killer robots. I picture the Marines I know patrolling Falluja with a heads-up display like HoloLens, tied to sensors and to an A.I. system that can process data faster and more precisely than humanly possible — an interface that helps them identify an object as a shoe, or an approaching truck as too light to be laden with explosives.Dear Tech Workers, U.S. Service Members Need Your Help
I enjoyed this interview and especially the title: “Humans Don’t Realize How Biased They Are Until AI Reproduces the Same Bias, Says UNESCO AI Chair.”
What are some core problems or research areas you want to approach?
People are now solving problems just by throwing an enormous amount of computation and data at them and trying every possible way. You can afford to do that if you are a big company and have a lot of resources, but people in developing countries cannot afford the data or the computational resources. So the theoretical challenge, or the fundamental challenge, is how to develop methods that are better understood and therefore don’t need experiments with hundreds of variants to get things to work.
Another thing is that some of the problems with current datasets, especially in terms of the usefulness of these systems for different cultures, is that there is a cultural bias in the data that has been collected. It is Western data informed with the Western way of seeing and doing things, so to some extent having data from different cultures and different environments is going to help make things more useful. You need to learn from data that is more relevant to the task.Humans Don’t Realize How Biased They Are Until AI Reproduces the Same Bias, Says UNESCO AI Chair
And of course:
“Solving” is probably too strong, but for addressing those problems, as I’ve said, the problem is that we don’t realise that they are the reflections of our own problems. We don’t realise how biased we are until we see an AI reproduce the same bias, and we see that it’s biased.
I chuckle a bit when I hear about biased humans going over biased data in the hopes of creating unbiased data. Bias is a really hard problem, and it’s always going to be with us in one form or another. Education and awareness are the most important tools for addressing.
Benjamin Heinzerling writes in The Gradient that the “Clever Hans effect” is alive and well in natural language processing (NLP) deep learning models:
Of course, the problem of learners solving a task by learning the “wrong” thing has been known for a long time and is known as the Clever Hans effect, after the eponymous horse which appeared to be able to perform simple intellectual tasks, but in reality relied on involuntary cues given by its handler. Since the 1960s, versions of the tank anecdote tell of a neural network trained by the military to recognize tanks in images, but actually learning to recognize different levels of brightness due to one type of tank appearing only in bright photos and another type only in darker ones.
Less anecdotal, Viktoria Krakovna has collected a depressingly long list of agents following the letter, but not the spirit of their reward function, with such gems as a video game agent learning to die at the end of the first level, since repeating that easy level gives a higher score than dying early in the harder second level. Two more recent, but already infamous cases are an image classifier claimed to be able to distinguish faces of criminals from those of law-abiding citizens, but actually recognizing smiles and a supposed “sexual orientation detector” which can be better explained as a detector of glasses, beards and eyeshadow.NLP’s Clever Hans Moment has Arrived
BERT is a fantastic NLP model, but it’s not displaying deep understanding of the material. For certain tasks, at least, it is exploiting statistical correlations better than you can. And that makes it hard to see what it’s doing.
Reminds me of one of my favorite quotes: “The first principle is that you must not fool yourself – and you are the easiest person to fool.” Richard Feynman
Anil Ananthaswamy, writing for Scientific American:
[T]the signals being collected by LIGO must be matched by supercomputers against hundreds of thousands of templates of possible gravitational-wave signatures. Promising signals trigger an internal alert; those that survive additional scrutiny trigger a public alert so that the global astronomy community can look for electromagnetic and neutrino counterparts.
Template matching is so computationally intensive that, for gravitational waves produced by mergers, astronomers use only four attributes of the colliding cosmic objects (the masses of both and the magnitudes of their spins) to make detections in real time. From there, LIGO scientists spend hours, days or even weeks performing more processing offline to further refine the understanding of a signal’s sources, a task called parameter estimation.
Seeking ways to make that labyrinthine process faster and more computationally efficient, in work published in 2018, Huerta and his research group at NCSA turned to machine learning. Specifically, Huerta and his then graduate student Daniel George pioneered the use of so-called convolutional neural networks (CNNs), which are a type of deep-learning algorithm, to detect and decipher gravitational-wave signals in real time.Faced with a Data Deluge, Astronomers Turn to Automation
And they learned a bit about what the neural networks are seeing:
For Ntampaka, these results suggest that machine-learning systems are not entirely immune to interpretation. “It’s a misunderstanding within the community that they only can be black boxes,” she says. “I think interpretability is on the horizon. It’s coming. We are starting to be able to do it now.” But she also acknowledges that had her team not already known the underlying physics connecting the x-ray emissions from galaxy clusters to their mass, it might not have figured out that the neural network was excising the cores from its analysis.
Gary Marcus, NYU professor of psychology and neural science, is skeptical in light of DeepMind’s loss of $572M last year:
My own guess?
Ten years from now we will conclude that deep reinforcement learning was overrated in the late 2010s, and that many other important research avenues were neglected. Every dollar invested in reinforcement learning is a dollar not invested somewhere else, at a time when, for example, insights from the human cognitive sciences might yield valuable clues. Researchers in machine learning now often ask, “How can machines optimize complex problems using massive amounts of data?” We might also ask, “How do children acquire language and come to understand the world, using less power and data than current AI systems do?” If we spent more time, money, and energy on the latter question than the former, we might get to artificial general intelligence a lot sooner.DEEPMIND’S LOSSES AND THE FUTURE OF ARTIFICIAL INTELLIGENCE
Deep learning has been so hyped that it will be difficult to meet expectations. And reinforcement learning has serious challenges when applied to real-world environments. But they are both revolutions in AI and will alter computing forever.
Peter Thiel in a NYT op-ed:
A.I.’s military power is the simple reason that the recent behavior of America’s leading software company, Google — starting an A.I. lab in China while ending an A.I. contract with the Pentagon — is shocking. As President Barack Obama’s defense secretary Ash Carter pointed out last month, “If you’re working in China, you don’t know whether you’re working on a project for the military or not.”Good for Google, Bad for America
And he’s not wrong. But it’s also not possible to prevent China from ultimately obtaining this technology.
Thiel is correct in the short-term, but also dangerously short-sighted. What’s the plan here? Further isolation and an arms race? Liberal democracies need to be focused on global frameworks (rule of law, free speech, free trade, free movement of people and information) that prevent war and human misery. This is an opportunistic easy rhetorical point, not a strategy.
Toby Walsh is an Australian professor of computer science working to prevent the development of autonomous robotic weapons:
[Y]ou can’t have machines deciding whether humans live or die. It crosses new territory. Machines don’t have our moral compass, our compassion and our emotions. Machines are not moral beings.
The technical argument is that these are potentially weapons of mass destruction, and the international community has thus far banned all other weapons of mass destruction.Toby Walsh, A.I. Expert, Is Racing to Stop the Killer Robots
Different emphasis, but again the focus is on human-in-the-loop safety.
An artist working for a European Commission project called SHERPA, which investigates the way “smart information systems” impact human rights, has built an AI-enabled water gun:
‘Our artist has built a water gun with a face recognition on it so it will only squirt water at women or it can be changed to recognise a single individual or people of a certain age,’ said Prof. Stahl. ‘The idea is to get people to think about what this sort of technology can do.’Getting AI ethics wrong could ‘annihilate technical progress’
The project is intended to highlight how biased or inaccurate AI systems can impact ordinary people. Does it think you’re between 30 and 50 years-old? Squirt.
AI’s generate fake news and AI’s detect them:
Grover is a strong detector of neural fake news precisely because it is simultaneously a state-of-the-art generator of neural fake news.Why We Released Grover
Grover can both generate and detect AI text. The blog post explains why the research team from the University of Washington decided to release the Grover model despite OpenAI’s decision that GPT-2 (a similarly powerful text generation model) was “too dangerous to release.” They conclude that the real danger is in even bigger models, and that research must continue.
OpenAI, on the heels of switching from a non-profit to a “capped profit,” announces a major investment:
Microsoft is investing $1 billion in OpenAI to support us building artificial general intelligence (AGI) with widely distributed economic benefits. We’re partnering to develop a hardware and software platform within Microsoft Azure which will scale to AGI. We’ll jointly develop new Azure AI supercomputing technologies, and Microsoft will become our exclusive cloud provider . . . .Microsoft invests in and partners with OpenAI to support us building beneficial AGI
The deal involves Microsoft becoming the exclusive cloud computing partner for OpenAI, so this is more like a partnership than a straight investment.