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
[google_news title=”” keyword=”artificial intelligence systems” num_posts=”10″ blurb_length=”0″ show_thumb=”left”]
via Google News
The Latest on: Artificial intelligence systems
- INSIDER: This AI-driven security system gives you the ability to ‘deter and mark intruders’on April 27, 2024 at 5:54 am
A Slovenia-based startup has created an artificial intelligence-driven security system that literally fights back against intruders.
- How Artificial Intelligence Is Changing The Way You Buy Travel Insuranceon April 27, 2024 at 3:52 am
Artificial intelligence is transforming the way you buy travel insurance. Here are the surprising ways it could affect your next vacation.
- Pope to bring his call for ethical artificial intelligence to G7 summit in June in southern Italyon April 27, 2024 at 3:01 am
Pope Francis is taking his call for artificial intelligence to be developed and used according to ethical lines to the Group of 7 industrialized nations.Italian Prime Minister Giorgia Meloni ...
- Artificial Intelligenceon April 26, 2024 at 5:14 am
Much as ChatGPT generates poetry, a new A.I. system devises blueprints for ... fueled by its services in generative artificial intelligence. Along with the higher spending, the owner of Facebook ...
- 2 Artificial Intelligence (AI) Stocks That Could Make You a Millionaireon April 26, 2024 at 2:20 am
UiPath ( PATH 0.26%) makes robotic process automation (RPA) software. This gives its clients the tools they need to automate repetitive tasks. UiPath also integrates artificial intelligence (AI) into ...
- A Once-in-a-Decade Investment Opportunity: 2 Artificial Intelligence (AI) Stocks to Buy Now and Hold Long Termon April 26, 2024 at 2:00 am
These stocks could help investors turn a profit as artificial intelligence takes root across business processes and consumer products.
- This Spectacular Vanguard ETF Could Turn $250 per Month Into $873,700, With Help From Artificial Intelligence (AI) Stocks Nvidia and Microsofton April 25, 2024 at 1:12 am
This Vanguard ETF tracks hundreds stocks in the information technology sector, the only market sector to outperform the S&P 500 over the past decade.
- Randomised controlled trials evaluating artificial intelligence in clinical practice: a scoping reviewon April 24, 2024 at 5:39 pm
This scoping review of randomised controlled trials on artificial intelligence (AI) in clinical practice reveals an expanding interest in AI across clinical specialties and locations. The USA and ...
- Artificial intelligence helps scientists engineer plants to fight climate changeon April 24, 2024 at 1:42 pm
The Intergovernmental Panel on Climate Change (IPCC) has declared that removing carbon from the atmosphere is now essential to fighting climate change and limiting global temperature rise. To support ...
- Emulating neurodegeneration and aging in artificial intelligence systemson April 24, 2024 at 3:30 am
In recent years, developers have introduced artificial intelligence (AI) systems that can simulate or reproduce various human abilities, such as recognizing objects in images, answering questions, and ...
via Bing News