Researchers have developed a new framework for deep neural networks that allows artificial intelligence (AI) systems to better learn new tasks while “forgetting” less of what they have learned regarding previous tasks.
The researchers 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.
“People are capable of continual learning; we learn new tasks all the time, without forgetting what we already know,” says Tianfu Wu, an assistant professor of electrical and computer engineering at NC State and co-author of a paper on the work. “To date, AI systems using deep neural networks have not been very good at this.”
“Deep neural network AI systems are designed for learning narrow tasks,” says 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.”
“We have proposed a new framework for continual learning, which decouples network structure learning and model parameter learning,” says Yingbo Zhou, co-lead author of the paper and a research scientist at Salesforce Research. “We call it the Learn to Grow framework. In experimental testing, we’ve found that it outperforms previous approaches to continual learning.”
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 datasets, 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 says. “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,” says 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.”
The Latest on: Continual learning for artificial intelligence
[google_news title=”” keyword=”continual learning for artificial intelligence” num_posts=”10″ blurb_length=”0″ show_thumb=”left”]
via Google News
The Latest on: Continual learning for artificial intelligence
- Software Defines Tacticson January 23, 2023 at 12:45 am
The book “The Bomber Mafia” revolves around the efforts of military officers to develop effective tactics for a bomber force, emphasizing the feedback ...
- Reskilling, upskilling, continuous learning: Three mantras imperative for both employers and employeeson January 23, 2023 at 12:25 am
Skilling employees and clients has always been an important area of focus for the top IT firms in India Infosys. Employees are expected to augment their skills regularly so they can excel in their ...
- AI Is Improving Faster Than Most Humans Realizeon January 22, 2023 at 11:44 pm
Technologies such as ChatGPT will profoundly change the economy, but they are getting better so quickly that it’s impossible to say how.
- How The Shift To Adult Learning Is Changing Businesson January 22, 2023 at 4:00 am
Adult learning creates stronger employees who, in turn, create more successful companies. It’s a shift into high gear that will move businesses that embrace it into the lead.
- Styles of machine learning: Intro to neural networkson January 19, 2023 at 4:10 am
Artificial intelligence (AI) is one of the most important and long-lived areas of research in computer science. It’s a broad area with crossover into philosophical questions about the nature of ...
- The Top Trends in Artificial Intelligence in 2023on January 18, 2023 at 4:00 pm
Artificial intelligence (AI) technologies have become dramatically ... of data points from diverse datasets that can be best analyzed and processed with machine learning tools. We’ve already seen many ...
- Artificial Intelligence Can Help Fight Wildfires: World Economic Forumon January 16, 2023 at 9:20 am
The average annual global cost of wildfires is around USD 50 billion, the World Economic Forum said on Monday.
- How can artificial intelligence protect whales? Offshore wind company has a planon January 12, 2023 at 11:05 pm
Vineyard Wind will partner with Charles River Analytics to use artificial intelligence to protect whales from devastating, often fatal vessel strikes.
- Growth of Artificial Intelligence in Pharma Manufacturingon January 12, 2023 at 8:02 pm
onza describes how artificial intelligence, machine learning, and big data are improving ... and sustainability while lowering costs. While a continuing increase in AI and ML integration will ...
- Quantum machine learning (QML) poised to make a leap in 2023on January 12, 2023 at 4:47 pm
Quantum machine learning (QML) combines quantum computing's power with ML's predictive capabilities to overcome classical algorithms' limitations.
via Bing News