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
  • Machine Learning

Predicting How Well Neural Networks Will Scale

  • February 27, 2020
  • relay

For all the progress researchers have made with machine learning in helping us doing things like crunch numbers, drive cars and detect cancer, we rarely think about how energy-intensive it is to maintain the massive data centers that make such work possible. Indeed, a 2017 study predicted that, by 2025, internet-connected devices would be using 20 percent of the world’s electricity.

ML Generalize

The inefficiency of machine learning is partly a function of how such systems are created. Neural networks are typically developed by generating an initial model, tweaking a few parameters, trying it again, and then rinsing and repeating. But this approach means that significant time, energy and computing resources are spent on a project before anyone knows if it will actually work.

MIT graduate student Jonathan Rosenfeld likens it to the 17th century scientists seeking to understand gravity and the motion of planets. He says that the way we develop machine learning systems today – in the absence of such understandings – has limited predictive power and is thus very inefficient.

“There still isn’t a unified way to predict how well a neural network will perform given certain factors like the shape of the model or the amount of data it’s been trained on,” says Rosenfeld, who recently developed a new framework on the topic with colleagues at MIT’s Computer Science and Artificial Intelligence Lab (CSAIL). “We wanted to explore whether we could move machine learning forward by trying to understand the different relationships that affect the accuracy of a network.”

The CSAIL team’s new framework looks at a given algorithm at a smaller scale, and, based on factors like its shape, can predict how well it will perform on a larger scale. This allows a data scientist  to determine if it’s worth continuing to devote more resources to train the system further.

Read More  Artificial Neural Networks Model Face Processing In Autism

“Our approach tells us things like the amount of data needed for an architecture to deliver a specific target performance, or the most computationally efficient trade-off between data and model size,” says MIT professor Nir Shavit, who co-wrote the new paper with Rosenfeld, former PhD student Yonatan Belinkov and Amir Rosenfeld of York University. “We view these findings as having far-reaching implications in the field by allowing researchers in academia and industry to better understand the relationships between the different factors that have to be weighed when developing deep learning models, and to do so with the limited computational resources available to academics.”

The framework allowed researchers to accurately predict performance at the large model and data scales using fifty times less computational power.

The aspect of deep learning performance that the team focused on is the so-called “generalization error,” which refers to the error generated when an algorithm is tested on real-world data. The team utilized the concept of model scaling, which involves changing the model shape in specific ways to see its effect on the error.

As a next step, the team plans to explore the underlying theories of what makes a specific algorithm’s performance succeed or fail. This includes experimenting with other factors that may impact the training of deep learning models.

The researchers’ work was funded in part by the Air Force Office of Scientific Research, the China-Singapore Suzhou Industrial Park, the Eli and Dorothy Berman Fellowship, the Harvard Mind Brain Behavior Initiative, and the National Science Foundation.

relay

Related Topics
  • CSAIL
  • MIT
  • MIT CSAIL
  • Neural Networks
You May Also Like
View Post
  • Artificial Intelligence
  • Technology
  • Tools

Ditching Google: The 3 Search Engines That Use AI To Give Results That Are Meaningful

  • March 23, 2023
View Post
  • Engineering
  • Machine Learning

Peacock: Tackling ML Challenges By Accelerating Skills

  • March 23, 2023
View Post
  • Data
  • Machine Learning
  • Platforms

Coop Reduces Food Waste By Forecasting With Google’s AI And Data Cloud

  • March 23, 2023
View Post
  • Artificial Intelligence
  • Machine Learning
  • Robotics

Gods In The Machine? The Rise Of Artificial Intelligence May Result In New Religions

  • March 23, 2023
View Post
  • Artificial Intelligence
  • Machine Learning

6 ways Google AI Is Helping You Sleep Better

  • March 21, 2023
View Post
  • Artificial Intelligence
  • Machine Learning

AI Could Make More Work For Us, Instead Of Simplifying Our Lives

  • March 21, 2023
View Post
  • Artificial Intelligence
  • Platforms

Microsoft To Showcase Purpose-Built AI Infrastructure At NVIDIA GTC

  • March 21, 2023
View Post
  • Artificial Intelligence
  • Engineering
  • Tools

The Next Generation Of AI For Developers And Google Workspace

  • March 21, 2023

Leave a Reply

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

Stay Connected!
LATEST
  • 1
    Ditching Google: The 3 Search Engines That Use AI To Give Results That Are Meaningful
    • March 23, 2023
  • 2
    Peacock: Tackling ML Challenges By Accelerating Skills
    • March 23, 2023
  • 3
    Coop Reduces Food Waste By Forecasting With Google’s AI And Data Cloud
    • March 23, 2023
  • 4
    Gods In The Machine? The Rise Of Artificial Intelligence May Result In New Religions
    • March 23, 2023
  • 5
    The Technology Behind A Perfect Cup Of Coffee
    • March 22, 2023
  • 6
    BigQuery Under The Hood: Behind The Serverless Storage And Query Optimizations That Supercharge Performance
    • March 22, 2023
  • 7
    6 ways Google AI Is Helping You Sleep Better
    • March 21, 2023
  • 8
    AI Could Make More Work For Us, Instead Of Simplifying Our Lives
    • March 21, 2023
  • 9
    Microsoft To Showcase Purpose-Built AI Infrastructure At NVIDIA GTC
    • March 21, 2023
  • 10
    The Next Generation Of AI For Developers And Google Workspace
    • March 21, 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
    ABB To Expand Robotics Factory In US
    • March 16, 2023
  • 2
    Introducing Microsoft 365 Copilot: Your Copilot For Work
    • March 16, 2023
  • 3
    Linux Foundation Training & Certification & Cloud Native Computing Foundation Partner With Corise To Prepare 50,000 Professionals For The Certified Kubernetes Administrator Exam
    • March 16, 2023
  • 4
    Intel Contributes AI Acceleration to PyTorch 2.0
    • March 15, 2023
  • 5
    Sumitovant More Than Doubles Its Research Output In Its Quest To Save Lives
    • March 21, 2023
  • /
  • Artificial Intelligence
  • Machine Learning
  • Robotics
  • Engineering
  • About

Input your search keywords and press Enter.