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
via Google News
The Latest on: Continual learning for artificial intelligence
- Litmus Partners With ProcessMiner to Offer Leading Edge Computing and Artificial Intelligence Platforms for Manufacturingon July 2, 2020 at 3:31 pm
Under terms of the agreement, both organizations will promote their respective platform capabilities throughout the manufacturing industry.
- Domino’s Australia Is Continuing Its AI Partnership - A Strong Vote Of Confidence For The Technologyon July 1, 2020 at 6:04 am
Domino’s Australia added a DOM Pizza Checker which leverages AI to scan each pizza to confirm they measure up to quality standards. Quality scores have increased by double digits since the rollout.
- Deliver More Effective Threat Intelligence with Federated Machine Learningon June 24, 2020 at 5:00 am
Cybercriminals never stop innovating. Their increased use of automated and scripted attacks that increase speed and scale makes them more sophisticated ...
- Open-source conversational artificial intelligence startup Rasa raises $26Mon June 23, 2020 at 8:05 pm
Founded in 2016, Rasa offers an infrastructure layer for conversation AI including the tools required to build contextual assistants. While offering the core platform on an open-source basis for free ...
- The Universal Knowledge Store on the Way to Artificial General Intelligenceon June 23, 2020 at 11:43 am
When you hear the term “Artificial Intelligence or AI,” it really refers to narrow AI—a system which may have superhuman “mental” abilities, but only in a narrow area of expertise. AIs now have expert ...
- ProcessMiner Partners with Litmus to Offer Leading Edge Computing and Artificial Intelligence Platforms for Manufacturingon June 23, 2020 at 11:00 am
ProcessMiner, an artificial intelligence platform for manufacturing, and Litmus, the Intelligent Edge Computing Platform for ...
- VirusTotal Adds Cynet's Artificial Intelligence-Based Malware Detectionon June 23, 2020 at 4:14 am
Google's multi-antivirus scanning service VirusTotal adds Cynet artificial intelligence-based malware detection ...
- Artificial Intelligence in Healthcare Market Overview, Top Key Players, Industry Growth Analysis, Forecast 2027on June 22, 2020 at 12:16 am
On the basis of offering, the global artificial intelligence in healthcare market is segmented into hardware, software, ...
- How Decentralization Could Alleviate Data Biases In Artificial Intelligenceon June 19, 2020 at 9:43 am
Data quality concerns continue to plague artificial intelligence. But blockchain-based incentive mechanisms could significantly change that.
- SOSi Invests in AppTek to Advance Artificial Intelligence and Machine Learning for Its Speech Recognition and Translation Offeringson June 17, 2020 at 12:06 pm
SOS International LLC (SOSi) announced today that its owners acquired a non-controlling interest in Applications Technology (AppTek), LLC, a leader in ...
via Bing News