“Hey Siri, how’s my hair?”
Your smartphone may soon be able to give you an honest answer, thanks to a new machine learning algorithm designed by U of T Engineering researchers Parham Aarabi (ECE) and Wenzhi Guo (ECE MASc 1T5).
The team designed an algorithm that learns directly from human instructions, rather than an existing set of examples, and outperformed conventional methods of training neural networks by 160 per cent. But more surprisingly, their algorithm also outperformed its own training by nine per cent — it learned to recognize hair in pictures with greater reliability than that enabled by the training, marking a significant leap forward for artificial intelligence.
Aarabi and Guo trained their algorithm to identify people’s hair in photographs — a much more challenging task for computers than it is for humans.
“Our algorithm learned to correctly classify difficult, borderline cases — distinguishing the texture of hair versus the texture of the background,” says Aarabi. “What we saw was like a teacher instructing a child, and the child learning beyond what the teacher taught her initially.”
Humans “teach” neural networks — computer networks that learn dynamically — by providing a set of labeled data and asking the neural network to make decisions based on the samples it’s seen. For example, you could train a neural network to identify sky in a photograph by showing it hundreds of pictures with the sky labeled.
This algorithm is different: it learns directly from human trainers. With this model, called heuristic training, humans provide direct instructions that are used to pre-classify training samples rather than a set of fixed examples. Trainers program the algorithm with guidelines such as “Sky is likely to be varying shades of blue,” and “Pixels near the top of the image are more likely to be sky than pixels at the bottom.”
Their work is published in the journal IEEE Transactions on Neural Networks and Learning Systems.
This heuristic training approach holds considerable promise for addressing one of the biggest challenges for neural networks: making correct classifications of previously unknown or unlabeled data. This is crucial for applying machine learning to new situations, such as correctly identifying cancerous tissues for medical diagnostics, or classifying all the objects surrounding and approaching a self-driving car.
“Applying heuristic training to hair segmentation is just a start,” says Guo. “We’re keen to apply our method to other fields and a range of applications, from medicine to transportation.”
The Latest on: AI algorithm heuristic training
via Google News
The Latest on: AI algorithm heuristic training
- The quest for explainable AIon May 6, 2022 at 11:01 pm
Explainable AI seeks to infuse algorithms with transparency so that even lay users can ensure their AI is operating within acceptable bounds.
- AI not immune to hacking, says Hon Hai Research Institute CEOon May 3, 2022 at 11:24 pm
AI has permeated well into the modern society playing a crucial role in a wide variety of applications, but AI systems, just like all kinds of computer operating systems, can hardly avoid being hacked ...
- IBM’s AutoAI Has The Smarts To Make Data Scientists A Lot More Productive – But What’s Scary Is That It’s Getting A Whole Lot Smarteron May 1, 2022 at 5:01 pm
I recently had the opportunity to discuss current IBM artificial intelligence developments with Dr. Lisa Amini, an IBM Distinguished Engineer and the Director of IBM Research Cambridge, home to ...
- Google Announces AI-Generated Summaries for Google Docson April 18, 2022 at 5:00 pm
The summarization is powered by a natural language processing (NLP) AI model based on the Transformer ... instead of simply masking random sentences, the training algorithm uses a heuristic ...
- From Language to Knowledge: Intel's path to cognitive AIon April 4, 2022 at 10:16 am
Today's AI has a workable mastery of natural language processing, image recognition, optimal path analysis, and other workhorse processes that can be applied to statistical and heuristic-based ...
- CSE 486/586 Introduction to Artificial Intelligence (3 credits)on March 31, 2022 at 8:28 pm
and the nearest neighbor classifier Describe and apply AI techniques in areas such as: evolutionary computation, neural networks, reinforcement learning, semantic web, fuzzy set theory, uncertainty, ...
- Graduate Certificate in Artificial Intelligence and Machine Learningon March 18, 2021 at 6:39 am
By combining foundational knowledge, applied training, and computational problem-solving knowledge, the AI and Machine Learning certificate program provides students with core knowledge in algorithms ...
- AI Learns Things That Humans Didn’t Teach Iton November 22, 2016 at 8:36 am
Most AI systems learn to do things ... a model called “heuristic training.” For instance, a typical instruction or algorithm guideline to identify water would go something like, “Sea ...
via Bing News