Liwaiwai Liwaiwai
  • /
  • Artificial Intelligence
  • Machine Learning
  • Robotics
  • Engineering
    • Architecture
    • Design
    • Software
    • Hybrid Cloud
    • Data
  • About
Liwaiwai Liwaiwai
  • /
  • Artificial Intelligence
  • Machine Learning
  • Robotics
  • Engineering
    • Architecture
    • Design
    • Software
    • Hybrid Cloud
    • Data
  • About
  • Artificial Intelligence

Shrinking Massive Neural Networks Used To Model Language

  • December 7, 2020
  • relay

You don’t need a sledgehammer to crack a nut.

Jonathan Frankle is researching artificial intelligence — not noshing pistachios — but the same philosophy applies to his “lottery ticket hypothesis.” It posits that, hidden within massive neural networks, leaner subnetworks can complete the same task more efficiently. The trick is finding those “lucky” subnetworks, dubbed winning lottery tickets.

In a new paper, Frankle and colleagues discovered such subnetworks lurking within BERT, a state-of-the-art neural network approach to natural language processing (NLP). As a branch of artificial intelligence, NLP aims to decipher and analyze human language, with applications like predictive text generation or online chatbots. In computational terms, BERT is bulky, typically demanding supercomputing power unavailable to most users. Access to BERT’s winning lottery ticket could level the playing field, potentially allowing more users to develop effective NLP tools on a smartphone — no sledgehammer needed.

“We’re hitting the point where we’re going to have to make these models leaner and more efficient,” says Frankle, adding that this advance could one day “reduce barriers to entry” for NLP.

Frankle, a PhD student in Michael Carbin’s group at the MIT Computer Science and Artificial Intelligence Laboratory, co-authored the study, which will be presented next month at the Conference on Neural Information Processing Systems. Tianlong Chen of the University of Texas at Austin is the lead author of the paper, which included collaborators Zhangyang Wang, also of UT Austin, as well as Shiyu Chang, Sijia Liu, and Yang Zhang, all of the MIT-IBM Watson AI Lab.

You’ve probably interacted with a BERT network today. It’s one of the technologies that underlies Google’s search engine, and it has sparked excitement among researchers since Google released BERT in 2018. BERT is a method of creating neural networks — algorithms that use layered nodes, or “neurons,” to learn to perform a task through training on numerous examples. BERT is trained by repeatedly attempting to fill in words left out of a passage of writing, and its power lies in the gargantuan size of this initial training dataset. Users can then fine-tune BERT’s neural network to a particular task, like building a customer-service chatbot. But wrangling BERT takes a ton of processing power.

“A standard BERT model these days — the garden variety — has 340 million parameters,” says Frankle, adding that the number can reach 1 billion. Fine-tuning such a massive network can require a supercomputer. “This is just obscenely expensive. This is way beyond the computing capability of you or me.”

Chen agrees. Despite BERT’s burst in popularity, such models “suffer from enormous network size,” he says. Luckily, “the lottery ticket hypothesis seems to be a solution.”

Read More  More Transparency And Understanding Into Machine Behaviors

To cut computing costs, Chen and colleagues sought to pinpoint a smaller model concealed within BERT. They experimented by iteratively pruning parameters from the full BERT network, then comparing the new subnetwork’s performance to that of the original BERT model. They ran this comparison for a range of NLP tasks, from answering questions to filling the blank word in a sentence.

The researchers found successful subnetworks that were 40 to 90 percent slimmer than the initial BERT model, depending on the task. Plus, they were able to identify those winning lottery tickets before running any task-specific fine-tuning — a finding that could further minimize computing costs for NLP. In some cases, a subnetwork picked for one task could be repurposed for another, though Frankle notes this transferability wasn’t universal. Still, Frankle is more than happy with the group’s results.

“I was kind of shocked this even worked,” he says. “It’s not something that I took for granted. I was expecting a much messier result than we got.”

This discovery of a winning ticket in a BERT model is “convincing,” according to Ari Morcos, a scientist at Facebook AI Research. “These models are becoming increasingly widespread,” says Morcos. “So it’s important to understand whether the lottery ticket hypothesis holds.” He adds that the finding could allow BERT-like models to run using far less computing power, “which could be very impactful given that these extremely large models are currently very costly to run.”

Frankle agrees. He hopes this work can make BERT more accessible, because it bucks the trend of ever-growing NLP models. “I don’t know how much bigger we can go using these supercomputer-style computations,” he says. “We’re going to have to reduce the barrier to entry.” Identifying a lean, lottery-winning subnetwork does just that — allowing developers who lack the computing muscle of Google or Facebook to still perform cutting-edge NLP. “The hope is that this will lower the cost, that this will make it more accessible to everyone … to the little guys who just have a laptop,” says Frankle. “To me that’s really exciting.”

Read More  Artificial Neural Networks Model Face Processing In Autism

This research was funded, in part, by IBM.

Source: CSAIL by Jonathan Frankle | Michael Carbin

relay

Related Topics
  • BERT
  • CSAIL
  • MIT
  • MIT Computer Science and Artificial Intelligence Laboratory
  • NLP
  • Research
You May Also Like
View Post
  • Artificial Intelligence
  • Technology

Unlocking The Secrets Of ChatGPT: Tips And Tricks For Optimizing Your AI Prompts

  • March 29, 2023
View Post
  • Artificial Intelligence
  • Technology

Try Bard And Share Your Feedback

  • March 29, 2023
View Post
  • Artificial Intelligence
  • Data
  • Data Science
  • Machine Learning
  • Technology

Google Data Cloud & AI Summit : In Less Than 12 Hours From Now

  • March 29, 2023
View Post
  • Artificial Intelligence
  • Technology

Talking Cars: The Role Of Conversational AI In Shaping The Future Of Automobiles

  • March 28, 2023
View Post
  • Artificial Intelligence
  • Tools

Document AI Introduces Powerful New Custom Document Classifier To Automate Document Processing

  • March 28, 2023
View Post
  • Artificial Intelligence
  • Design
  • Practices

How AI Can Improve Digital Security

  • March 27, 2023
View Post
  • Artificial Intelligence
  • Machine Learning
  • Technology

ChatGPT 4.0 Finally Gets A Joke

  • March 27, 2023
View Post
  • Artificial Intelligence
  • Machine Learning
  • Technology

Mr. Cooper Is Improving The Home-buyer Experience With AI And ML

  • March 24, 2023

Leave a Reply

Your email address will not be published. Required fields are marked *

Stay Connected!
LATEST
  • 1
    Unlocking The Secrets Of ChatGPT: Tips And Tricks For Optimizing Your AI Prompts
    • March 29, 2023
  • 2
    Try Bard And Share Your Feedback
    • March 29, 2023
  • 3
    Google Data Cloud & AI Summit : In Less Than 12 Hours From Now
    • March 29, 2023
  • 4
    Talking Cars: The Role Of Conversational AI In Shaping The Future Of Automobiles
    • March 28, 2023
  • 5
    Document AI Introduces Powerful New Custom Document Classifier To Automate Document Processing
    • March 28, 2023
  • 6
    How AI Can Improve Digital Security
    • March 27, 2023
  • 7
    ChatGPT 4.0 Finally Gets A Joke
    • March 27, 2023
  • 8
    Mr. Cooper Is Improving The Home-buyer Experience With AI And ML
    • March 24, 2023
  • 9
    My First Pull Request At Age 14
    • March 24, 2023
  • 10
    The 5 Podcasts To Check If You Want To Get Up To Speed On AI
    • March 24, 2023

about
About
Hello World!

We are liwaiwai.com. Created by programmers for programmers.

Our site aims to provide materials, guides, programming how-tos, and resources relating to artificial intelligence, machine learning and the likes.

We would like to hear from you.

If you have any questions, enquiries or would like to sponsor content, kindly reach out to us at:

[email protected]

Live long & prosper!
Most Popular
  • 1
    GPT-4 : The Latest Milestone From OpenAI
    • March 24, 2023
  • 2
    Ditching Google: The 3 Search Engines That Use AI To Give Results That Are Meaningful
    • March 23, 2023
  • 3
    Peacock: Tackling ML Challenges By Accelerating Skills
    • March 23, 2023
  • 4
    Coop Reduces Food Waste By Forecasting With Google’s AI And Data Cloud
    • March 23, 2023
  • 5
    Gods In The Machine? The Rise Of Artificial Intelligence May Result In New Religions
    • March 23, 2023
  • /
  • Artificial Intelligence
  • Machine Learning
  • Robotics
  • Engineering
  • About

Input your search keywords and press Enter.