Lifelong-Deep Neural Network™
technology: the only AI that gets smarter after every use

Lifelong-DNN emulates the way biological brains see the world and continuously learns from it. L-DNN pushes AI beyond on-device inference and allows it to learn on the device itself, significantly reducing the amount of data needed and training time and, enables real-time learning on-device. Today Lifelong-DNN powers over 50M devices globally, ranging from cameras, to smart phones, drones, and industrial machines.

REQUEST DEMO
dnn-technology
technology-ai-brain

Designed for NASA and Mars, brought down to Earth for real-world applications.

Neurala designed Lifelong-DNN after years of intense R&D with NASA, and today it is protected by over 25 awarded and pending patents.

NASA asked Neurala to design a novel Artificial Intelligence algorithm able to power an autonomous robot navigating an uncharted environment. The AI needed not only to control the robot, emulating aspects of human perception, planning, and motor control using very little compute power, but it also needed to learn new information on-device, without relying on internet connectivity and external compute resources. Back then, traditional DNNs needed lots of compute power and time to learn new information and even today, traditional DNNs rely on this paradigm. L-DNN is the answer to the quest for an AI that can learn continuously, the way we do.

TALK WITH OUR AI EXPERTS

Brain Builder™: the AI platform that never stops learning.

Brain Builder incorporates Lifelong-DNN technology that enables rapid AI prototyping and deployment. This end to end approach allows users to annotate and train at the same time, using just a few images, instead of thousands. It offers instant feedback on performance, quickly iterating instead of having to wait hours or weeks to understand the results.

Brain Builder™: the AI platform that never stops learning.

VS.
Less Data Required
VS.
Instantaneous Learning
VS.
Directly on small compute edge
VS.
Continues to learn

L-DNN trains 10,000 times faster than traditional DNNs

Faster training gives significant advantages:

  • the amount of time and processing power required to augment the intelligence of the AI are negligible
  • every image can be used to augment the precision of the AI model
  • the precision gains are instantaneous, rather than having to wait for hours, days or weeks
Artifical Intelligence Concept Machine

AI learning at the Edge

L-DNN’s fast learning speed means learning can happen

  • directly on a small compute edge
  • without requiring an Internet connection
  • without needing to send data outside the edge to the cloud, so that data privacy and protection are maintained
Expert Clicks on Circuit

More accurate than traditional DNNs

The patented L-DNN technology emulates the fast learning humans and animals perform on every piece of ‘data’ they encounter. L-DNN accelerated learning speed means that new learning can occur as new data is available. L-DNN keeps learning. Traditional DNNs do not.

Talk with our AI Experts
Accuracy

Using only good product samples

Neurala’s AI can be trained with only GOOD image examples, unlike traditional DNNs that need both good and bad examples. Working with good samples only, L-DNN produces an anomaly signal when an alteration occurs. This results in tremendous time savings since defective products appear infrequently and randomly, and it would therefore take hours to update the network.

Aluminum Parts on Assembly Line

L-DNN: Rapid Prototyping and Real Time Learning

By merging L-DNN with proprietary implementations, Neurala has solved shortcomings with traditional deep learning architectures. This enables rapid prototyping in AI model development, as well as real-time learning in deployed solutions, saving both time and money.
Request demo

Award-winning technology powering over 50M devices

SearchEnterpriseAI Logo
SearchEnterpriseAI Logo
SearchEnterpriseAI Logo
SearchEnterpriseAI Logo
SearchEnterpriseAI Logo
SearchEnterpriseAI Logo

Hardware agnostic deployment with Brain Builder

Examples of Platform vs Hardware Applications

Cameras
cameras-icon

Camera with an Orange Pi Zero

(ARM CPU): 0.5-2 FPS

Police Body camera

(Adreno 330 GPU): 1-3 FPS

Web Camera

(NVIDIA TX1 GPU): 5-8 FPS

Mobile Phones
mobile-phones-icon

Samsung Note5

(Mali T760 GPU): 3-5 FPS

Samsung Galaxy S8+

(Adreno 540 GPU): 6-11 FPS

Mate 10

(Kirin 970 NPU): 11-18 F

Drones
drones-icon

Samsung Note5

(Mali T760 GPU): 3-5 FPS

Samsung Galaxy S8+

(Adreno 540 GPU): 6-11 FPS

Mate 10

(Kirin 970 NPU): 11-18 F

Servers
servers-icon

Laptop
(NVIDIA 1060 Ti GPU): 40-50 FPS

Workstation
(NVIDIA Tesla V100 GPU): 60-80 FPS

Request a demo