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
- New artificial intelligence technology detects strokes to cut time to treatment and save liveson May 27, 2022 at 12:13 pm
A lightning-fast system at Main Line Health reads scans the second they are done, finding blood clots that cause strokes right away to help save lives.
- Artificial intelligence helps in the identification of astronomical objectson May 27, 2022 at 6:54 am
Classifying celestial objects is a long-standing problem. With sources at near unimaginable distances, sometimes it's difficult for researchers to distinguish between objects such as stars, galaxies, ...
- Artificial General Intelligence: What's Behind You?on May 27, 2022 at 5:30 am
AGI is the hypothetical ability of an intelligent agent to understand or learn any intellectual task that a human being can.
- simpleshow Explainer Videos Powered by Artificial Intelligenceon May 26, 2022 at 8:36 pm
The fundamental premise of artificial intelligence is straightforward: a computer/machine analyzes a set of data fed into the system and solves issues. In this sense, it mimics the human brain. In ...
- ISU professor creates artificial intelligence that composes musicon May 26, 2022 at 3:00 pm
When it comes to music, composition is one problem that has always vexed computers. Sure, they can string a few notes or lyrics together, but creating motifs — ...
- Three Key Artificial Intelligence Adoption Pitfalls to Avoid in 2022 and Beyondon May 26, 2022 at 11:13 am
The avenue to adopting Artificial intelligence (AI) isn’t always straightforward. Business leaders must collect the appropriate data, identify the right technologies for their firm, and teach their ...
- After shooting, Oxford High turned to artificial intelligence to detect weaponson May 25, 2022 at 8:00 pm
Biometric scanners to detect weapons and window shields are also among the security measures Oxford Community Schools are evaluating.
- Automotive Artificial Intelligence Software Market by Global Size, Trends and Research Analysis 2022 predicted to reach USD 53.98 billion by 2030on May 25, 2022 at 1:30 am
Automotive Artificial Intelligence Software is a system which is used for the altertion, process and functions and others to imitate human mind and absorbs the problem solving capacity automatically.
- Artificial intelligence is breaking patent lawon May 23, 2022 at 5:00 pm
The patent system assumes that inventors are human. Inventions devised by machines require their own intellectual property law and an international treaty.
- Artificial Intelligence (AI) in Food and Beverage Market to enjoy explosive growth | Aboard Software, Analytical Flavor Systems, Deepnify, etc.on May 20, 2022 at 3:14 am
Impact Analysis on the Growth of Artificial Intelligence (AI) in Food and Beverage Market The global artificial intelligence in food & beverages ...
via Bing News