A project of the U.S. Army has developed a new framework for deep neural networks that allows artificial intelligence systems to better learn new tasks while forgetting less of what they have learned in previous tasks.
The North Carolina State University researchers, funded by the Army, have also demonstrated that using the framework to learn a new task can make the AI better at performing previous tasks, a phenomenon called backward transfer.
“The Army needs to be prepared to fight anywhere in the world so its intelligent systems also need to be prepared,” said Dr. Mary Anne Fields, program manager for Intelligent Systems at Army Research Office, an element of U.S. Army Combat Capabilities Development Command’s Army Research Lab. “We expect the Army’s intelligent systems to continually acquire new skills as they conduct missions on battlefields around the world without forgetting skills that have already been trained. For instance, while conducting an urban operation, a wheeled robot may learn new navigation parameters for dense urban cities, but it still needs to operate efficiently in a previously encountered environment like a forest.”
The research team proposed a new framework, called Learn to Grow, for continual learning, which decouples network structure learning and model parameter learning. In experimental testing it outperformed previous approaches to continual learning.
“Deep neural network AI systems are designed for learning narrow tasks,” said Xilai Li, a co-lead author of the paper and a Ph.D. candidate at NC State. “As a result, one of several things can happen when learning new tasks, systems can forget old tasks when learning new ones, which is called catastrophic forgetting. Systems can forget some of the things they knew about old tasks, while not learning to do new ones as well. Or systems can fix old tasks in place while adding new tasks — which limits improvement and quickly leads to an AI system that is too large to operate efficiently. Continual learning, also called lifelong-learning or learning-to-learn, is trying to address the issue.”
To understand the Learn to Grow framework, think of deep neural networks as a pipe filled with multiple layers. Raw data goes into the top of the pipe, and task outputs come out the bottom. Every “layer” in the pipe is a computation that manipulates the data in order to help the network accomplish its task, such as identifying objects in a digital image. There are multiple ways of arranging the layers in the pipe, which correspond to different “architectures” of the network.
When asking a deep neural network to learn a new task, the Learn to Grow framework begins by conducting something called an explicit neural architecture optimization via search. What this means is that as the network comes to each layer in its system, it can decide to do one of four things: skip the layer; use the layer in the same way that previous tasks used it; attach a lightweight adapter to the layer, which modifies it slightly; or create an entirely new layer.
This architecture optimization effectively lays out the best topology, or series of layers, needed to accomplish the new task. Once this is complete, the network uses the new topology to train itself on how to accomplish the task — just like any other deep learning AI system.
“We’ve run experiments using several data sets, and what we’ve found is that the more similar a new task is to previous tasks, the more overlap there is in terms of the existing layers that are kept to perform the new task,” Li said. “What is more interesting is that, with the optimized — or “learned” topology — a network trained to perform new tasks forgets very little of what it needed to perform the older tasks, even if the older tasks were not similar.”
The researchers also ran experiments comparing the Learn to Grow framework’s ability to learn new tasks to several other continual learning methods, and found that the Learn to Grow framework had better accuracy when completing new tasks.
To test how much each network may have forgotten when learning the new task, the researchers then tested each system’s accuracy at performing the older tasks — and the Learn to Grow framework again outperformed the other networks.
“In some cases, the Learn to Grow framework actually got better at performing the old tasks,” said Caiming Xiong, the research director of Salesforce Research and a co-author of the work. “This is called backward transfer, and occurs when you find that learning a new task makes you better at an old task. We see this in people all the time; not so much with AI.”
“This Army investment extends the current state of the art machine learning techniques that will guide our Army Research Laboratory researchers as they develop robotic applications, such as intelligent maneuver and learning to recognize novel objects,” Fields said. “This research brings AI a step closer to providing our warfighters with effective unmanned systems that can be deployed in the field.”
Learn more: Army-funded research boosts memory of AI systems
The Latest on: Artificial intelligence systems
via Google News
The Latest on: Artificial intelligence systems
- Box-Level Systems Marry AI and IoTon January 17, 2022 at 5:10 pm
Intelligence at the Edge We’re now entrenched in an era where artificial intelligence (AI) is easily done at an embedded systems level. That means system d ...
- Send in the clones: Using artificial intelligence to digitally replicate human voiceson January 17, 2022 at 1:18 pm
Thanks to advances in artificial intelligence, it's never been easier or more affordable to make a perfect facsimile of a human voice: a celebrity, a world leader or even a public-radio reporter.
- Researchers demonstrate multimodal transistor in artificial neural networkson January 17, 2022 at 6:23 am
Researchers at the University of Surrey report a proof-of-concept demonstration of a multimodal transistor (MMT) in artificial neural networks, which mimic the human brain.
- How Artificial Intelligence Affects Digital Marketingon January 16, 2022 at 1:35 am
Hundreds of sectors are being transformed by artificial intelligence. Artificial intelligence is transforming the way brands do business in a variety of industries, including finance, technology, and ...
- 3 Top Artificial Intelligence Stocks to Buy in Januaryon January 15, 2022 at 8:24 am
Nearly every business is using artificial intelligence in some form today, but these three companies have AI at their core. Upstart and Lemonade are both using AI to make decisions in industries that ...
- Global Artificial Intelligence (AI) in Call Centers Market to Reach US$3.5 Billion by the Year 2026on January 14, 2022 at 2:22 am
Abstract: Global Artificial Intelligence (AI) in Call Centers Market to Reach US$3. 5 Billion by the Year 2026 . Artificial intelligence is playing a vital role in ensuring transformation of call ...
- 3 Entrepreneurial Uses of Artificial Intelligence That Will Change Your Businesson January 13, 2022 at 8:00 am
Artificial intelligence has the power to transform your business. It can be used for everything from customer service to sales, and it's easy to implement as well. Artificial intelligence is already ...
- Artificial Intelligence in Security Market to See Booming Growth | Acalvio, Cylance, IBM, Micronon January 13, 2022 at 3:07 am
Advance Market Analytics published a new research publication on Artificial Intelligence in Security Market Insights to 2026 with 232 pages and enriched with self explained Tables and charts in ...
- How artificial intelligence could influence hospital triageon January 12, 2022 at 11:00 am
As the latest virus surge swamps hospitals, researchers consider the role machine learning could play in making decisions about care.
- Changing the Retail System: Artificial Intelligenceon January 8, 2022 at 6:29 am
Artificial intelligence solutions, which continue to influence the world and become widespread in various industries, continue to break new ground in the retail world. As a result, the retail sector ...
via Bing News