New system enables pattern-recognition systems to convey what they learn to humans.
Computers are good at identifying patterns in huge data sets. Humans, by contrast, are good at inferring patterns from just a few examples.
In a paper appearing at the Neural Information Processing Society’s conference next week, MIT researchers present a new system that bridges these two ways of processing information, so that humans and computers can collaborate to make better decisions.
The system learns to make judgments by crunching data but distills what it learns into simple examples. In experiments, human subjects using the system were more than 20 percent better at classification tasks than those using a similar system based on existing algorithms.
“In this work, we were looking at whether we could augment a machine-learning technique so that it supported people in performing recognition-primed decision-making,” says Julie Shah, an assistant professor of aeronautics and astronautics at MIT and a co-author on the new paper. “That’s the type of decision-making people do when they make tactical decisions — like in fire crews or field operations. When they’re presented with a new scenario, they don’t do search the way machines do. They try to match their current scenario with examples from their previous experience, and then they think, ‘OK, that worked in a previous scenario,’ and they adapt it to the new scenario.”
In particular, Shah and her colleagues — her student Been Kim, whose PhD thesis is the basis of the new paper, and Cynthia Rudin, an associate professor of statistics at the MIT Sloan School of Management — were trying to augment a type of machine learning known as “unsupervised.”
In supervised machine learning, a computer is fed a slew of training data that’s been labeled by humans and tries to find correlations — say, those visual features that occur most frequently in images labeled “car.” In unsupervised machine learning, on the other hand, the computer simply looks for commonalities in unstructured data. The result is a set of data clusters whose members are in some way related, but it may not be obvious how.
The most common example of unsupervised machine learning is what’s known as topic modeling, in which a system clusters documents together according to their most characteristic words. Since the data is unlabeled, the system can’t actually deduce the topics of the documents. But a human reviewing its output would conclude that, for instance, the documents typified by the words “jurisprudence” and “appellate” are legal documents, while those typified by “tonality” and “harmony” are music-theory papers.
The MIT researchers made two major modifications to the type of algorithm commonly used in unsupervised learning. The first is that the clustering was based not only on data items’ shared features, but also on their similarity to some representative example, which the researchers dubbed a “prototype.”
The other is that rather than simply ranking shared features according to importance, the way a topic-modeling algorithm might, the new algorithm tries to winnow the list of features down to a representative set, which the researchers dubbed a “subspace.” To that end, the algorithm imposes a penalty on subspaces that grow too large. So when it’s creating its data clusters, it has to balance three sometimes-competing objectives: similarity to prototype, subspace size, and clear demarcations between clusters.
“You have to pick a good prototype to describe a good subspace,” Kim explains. “At the same time, you have to pick the right subspace such that the prototype makes sense. So you’re doing it all simultaneously.”
The researchers’ first step was to test their new algorithm on a few classic machine-learning tasks, to make sure that the added constraints didn’t impair its performance. They found that on most tasks, it performed as well as its precursor, and on a few, it actually performed better. Shah believes that that could be because the prototype constraint prevents the algorithm from assembling feature lists that contain internal contradictions.
Suppose, for instance, that an unsupervised-learning algorithm was trying to characterize voters in a population. A plurality of the voters might be registered as Democrats, but a plurality of Republicans may have voted in the last primary. The conventional algorithm might then describe the typical voter as a registered Democrat who voted in the last Republican primary. The prototype constraint makes that kind of result very unlikely, since no single voter would match its characterization.
Next, the researchers conducted a set of experiments to determine whether prototype-based machine learning could actually improve human decision-making. Kim culled a set of recipes from an online database in which they had already been assigned categories — such as chili, pasta, and brownies — and distilled them to just their ingredient lists. Then she fed the lists to both a conventional topic-modeling algorithm and the new, prototype-constrained algorithm.
The Latest on: Computers that teach by example
via Google News
The Latest on: Computers that teach by example
- How Pandora built a successful apprenticeship programon September 8, 2019 at 2:29 pm
- How to become a machine learning engineeron September 8, 2019 at 11:06 am
For example ... the Turing test. Computer vision on the other hand – the ability of a program to identify objects in a scene – is accomplished via machine learning. By looking at hundreds of thousands ...
- Parents being extra cautious and teaching their children about personal safetyon September 6, 2019 at 5:01 pm
“And if children are playing games on their smartphones or computers, I think we as parents need to monitor ... it is important to educate young children on personal safety. She said she was teaching ...
- With the help of her students, U of T researcher and McLean Award winner is building the future of computer architectureon September 4, 2019 at 2:44 pm
Natalie Enright Jerger, a professor in U of T's Faculty of Applied Science & Engineering, is considered a worldwide expert in “on-chip” networks (photo by Roberta Baker) As a computer architect ... s ...
- Autism, through the eyes of a computeron September 4, 2019 at 4:06 am
Accompanying each adolescent boy in this classroom is a teaching assistant wearing thick ... maintaining his composure during shopping trips, for example — that he can exchange for time on his ...
- MIT developed a course to teach tweens about the ethics of AIon September 4, 2019 at 4:00 am
For example, students have to write ... will inevitably be part of a larger movement to teach more computational thinking, sometimes pitched as computer science, or coding. Payne believes that ...
- AI thinks this flood photo is a toilet. Fixing that could improve disaster response.on August 30, 2019 at 7:57 am
In one particularly egregious example, ImageNet, the golden standard for image classification ... details of on-the-ground conditions before their arrival. But most computer vision systems are trained ...
- Modern hunter-gatherer children could tell us how human culture evolved and inspire new ways of teachingon August 29, 2019 at 7:33 am
At the beginning, there was language and speech, then came the written word and printing, radio communication and telephone, then computers ... hunter-gatherer groups. Teaching is another great ...
- Students distracted by tech leave professors longing for eye contacton August 29, 2019 at 3:34 am
When I started my undergraduate education 21 years ago, computers were a luxury and talking to the professor ... As a university instructor with nine years of experience teaching undergraduate and ...
- Coughing, colonoscopies and porn: the tedium of teaching AIon August 28, 2019 at 8:56 pm
When she found one – they look a bit like a slimy, angry pimple – she marked it with her computer mouse and keyboard, drawing a digital circle around the tiny bulge. She was not trained as a doctor, ...
via Bing News