Predict your next investment

Preferred Networks company logo
Corporation
INTERNET | Internet Software & Services / Business Intelligence, Analytics & Performance Mgmt
preferred.jp

See what CB Insights has to offer

Founded Year

2014

Stage

Corporate Minority - VII | Alive

Total Raised

$152.19M

Last Raised

$9.22M | 2 yrs ago

About Preferred Networks

Preferred Networks (PFN) provides IoT-centric deep learning systems. The company advocates Edge Heavy Computing as a way to handle the enormous amount of data generated by devices in a distributed and collaborative manner at the edge of the network, with a focus on three business areas: transportation, manufacturing, and bio/healthcare. PFN develops and provides Chainer, an open-source deep learning framework.

Preferred Networks Headquarter Location

Otemachi Building 1-6-1 Otemachi, Chiyoda-ku

Tokyo, 100-0004,

Japan

Latest Preferred Networks News

Tokyo U & Preferred Networks Propose a Fast Estimation Method for the Stability of Ensemble Feature Selectors

Aug 11, 2021

A research team from Tokyo University and Preferred Networks proposes a fast simulation-based method for estimating the stability of ensemble selectors. Feature selection is a core concept in machine learning. Aiming at selecting a subset of relevant features for use in model construction, feature selection is a crucial step that can dramatically impact model performance. One of the most commonly used methods for improving the stability of feature selectors is to integrate the results of multiple feature selectors, aka ensemble feature selection. Drawbacks to this approach are that it is time-consuming, and, until now, there has been limited research on how to reduce its computational cost when estimating stability. To address these issues, a research team from Tokyo University and Preferred Networks has proposed a fast, simulation-based method for estimating the stability of ensemble feature selectors. The idea behind the proposed method is to build a feature selector simulator that mimics the behaviour of the base selector and uses a simulated ensemble feature selector to estimate the stability. The proposed algorithm constructs a set of simulated selectors that model the base selector as well as the dataset. It can then quickly calculate stability by creating simulated ensemble feature selectors that contain two parameters: the number of useful features for the task (n); and a probability that reflects the uncertainty derived from both feature selectors and the dataset (p). Because these two parameters are obtained by running the real selector, in this study the researchers assume that the parameters have already been estimated. In the proposed algorithm’s workflow, computational cost is dependant on the run trials of the real selectors, and so the overall computational complexity is relatively low, enabling a faster stability computation process. To demonstrate the applicability of their proposed method, the team conducted experiments on three microarray gene expression datasets: Colon (Ding and Peng, 2005), Lymphoma (Ding and Peng, 2005), and Prostate (Nie et al., 2010). For their base feature selector, they used a trained random forest that assigns an importance score to each feature and another random forest as a predictor for evaluating the performance of the selected features. They employed the pair-wise Jaccard similarity as their stability index. The results show that the proposed method can accurately estimate the stability of ensemble feature selectors while maintaining a low computation cost. The team believes their simulation method can aid in the evaluation of ensemble feature selection algorithms in terms of stability while saving time by reducing the required number of executions of the real feature selectors. The paper Fast Estimation Method for the Stability of Ensemble Feature Selectors is on arXiv . Author: Hecate He | Editor: Michael Sarazen, Chain Zhang We know you don’t want to miss any news or research breakthroughs. Subscribe to our popular newsletter  Synced Global AI Weekly  to get weekly AI updates. Share this:

Predict your next investment

The CB Insights tech market intelligence platform analyzes millions of data points on venture capital, startups, patents , partnerships and news mentions to help you see tomorrow's opportunities, today.

Research containing Preferred Networks

Get data-driven expert analysis from the CB Insights Intelligence Unit.

CB Insights Intelligence Analysts have mentioned Preferred Networks in 2 CB Insights research briefs, most recently on Jun 24, 2021.

Expert Collections containing Preferred Networks

Expert Collections are analyst-curated lists that highlight the companies you need to know in the most important technology spaces.

Preferred Networks is included in 6 Expert Collections, including IIOT Landscape.

I

IIOT Landscape

498 items

Companies in the industrial internet of things space, including sensor analytics platforms, edge computing, asset tracking, and more.

U

Unicorns- Billion Dollar Startups

832 items

A

AI 100 2018

99 items

C

Cloud Computing

1,330 items

Cloud computing startups develop technologies for remote (off-premises) servers used to store, manage, and process data.

E

Enterprise SaaS

2,452 items

Software-as-a-service (SaaS) – internet based software offered as a subscription – continues to become the de facto standard for software distribution and consumption. Enterprise SaaS continues to show particular promise, emerging as one of the most well-funded categories. Startu

I

Internet of Things ( IoT )

3,149 items

Preferred Networks Patents

Preferred Networks has filed 84 patents.

The 3 most popular patent topics include:

  • Machine learning
  • Artificial neural networks
  • Artificial intelligence
patents chart

Application Date

Grant Date

Title

Related Topics

Status

3/20/2018

9/21/2021

Muscles of the upper limb, Fingers, Hand, Horse anatomy, Upper limb anatomy

Grant

00/00/0000

00/00/0000

Subscribe to see more

Subscribe to see more

Subscribe to see more

00/00/0000

00/00/0000

Subscribe to see more

Subscribe to see more

Subscribe to see more

00/00/0000

00/00/0000

Subscribe to see more

Subscribe to see more

Subscribe to see more

00/00/0000

00/00/0000

Subscribe to see more

Subscribe to see more

Subscribe to see more

Application Date

3/20/2018

00/00/0000

00/00/0000

00/00/0000

00/00/0000

Grant Date

9/21/2021

00/00/0000

00/00/0000

00/00/0000

00/00/0000

Title

Subscribe to see more

Subscribe to see more

Subscribe to see more

Subscribe to see more

Related Topics

Muscles of the upper limb, Fingers, Hand, Horse anatomy, Upper limb anatomy

Subscribe to see more

Subscribe to see more

Subscribe to see more

Subscribe to see more

Status

Grant

Subscribe to see more

Subscribe to see more

Subscribe to see more

Subscribe to see more

Preferred Networks Web Traffic

Rank
Page Views per User (PVPU)
Page Views per Million (PVPM)
Reach per Million (RPM)
CBI Logo

Preferred Networks Rank

CB Insights uses Cookies

CBI websites generally use certain cookies to enable better interactions with our sites and services. Use of these cookies, which may be stored on your device, permits us to improve and customize your experience. You can read more about your cookie choices at our privacy policy here. By continuing to use this site you are consenting to these choices.