MIT taught a neural network how to show its work

Tech	2018-9-13 00:44

MIT’s Lincoln Laboratory Intelligence and Decision Technologies Group yesterday unveiled a neural network capable of explaining its reasoning. It’s the latest attack on the black box problem, and a new tool for combating biased AI.

Dubbed the Transparency by Design Network (TbD-net), MIT’s latest machine learning marvel is a neural network designed to answer complex questions about images. The network parses a query by breaking it down into subtasks that are handled by individual modules. If you asked it to determine the color of “the large square” in a picture showing several different shapes of varying size and color, for…

This story continues at The Next Web

. origin »

Santiment Network Token (SAN) на Currencies.ru

$ 0.1264 (+0.00%)

Объем 24H $0

Изменеия 24h: 5.76 %, 7d: 3.92 %

Cегодня L: $0.1264 - H: $0.1264

Капитализация $0 Rank 6408

Цена в час новости $ 0.4544 (-72.18%)

network neural mit latest handled individual subtasks

386 +

Источник: thenextweb.com

This neural network lets you build and evolve little AI creatures

The magic of artificial intelligence brings us better Netflix recommendations, better cancer diagnoses, and the ability to becomes gods by creating our own AI-powered creatures and forcing them to evolve for our amusement.

MIT’s depression-detecting AI might be its scariest creation yet

Researchers at the Massachusetts Institute of Technology (MIT) recently developed a neural network that predicts, with a relatively high level of accuracy, the likelihood a person has a cognitive impairment.

IBM came up with a watermark for neural networks

The creation and development of a robust neural network is a labor-intensive and time consuming endeavor. That’s why a team of IBM researchers recently developed a way for AI developers to protect their intellectual property.

Experts warn DeepFakes could influence 2020 US election

Fake AI-generated videos featuring political figures could be all the rage during the next election cycle, and that’s bad news for democracy. A recently released study indicates that DeepFakes, a neural network that creates fake videos of real people, represents one of the largest threats posed by artificial intelligence.

Facebook is teaching AI to talk you through directions

Facebook is developing a “Talk the Walk” AI capable of giving walking directions without knowing a user’s location. What it is: A team comprised of a researcher from the University of Montreal in Canada and Facebook Artificial Intelligence Research (FAIR) scientists recently published a white paper describing a neural network capable of giving a person plain language directions without the use of GPS or other location tracking aids.

AI Opinion ICO

AI Opinion is a neural network, which learns to detect most nosology types on various medical images, such as x-ray scans, ultrasound, computed tomography and magnetic resonance screening. The project will provide the opportunity to get the view of artificial intelligence on health state of a user.

Самое свежее на Currencies.ru

Interview: ‘This is just the tip of the iceberg,’ says BitMind founder Ken Jon Miyachi on deepfake scams

As deepfake technology becomes more sophisticated, so do the scams it enables. In just the first four months of 2025, 163 deepfake-related incidents were reported, leading to over $200 million in losses.

Interview: ‘No one, not even us, can see what you are doing’ — NymVPN’s CSO ahead of launch

Privacy-focused firm Nym, backed by a16z and other investors, has introduced its zero-knowledge payment system along with a perpetual token buyback model. Nym’s decentralized DePIN payments system and token buyback are now live for public testing.

10 Common FAQs Around Passphrases in Trezor Wallets

The official blog for Trezor, the world’s first hardware wallet. All you need to keep your digital assets safe. - Medium

Solana Rolls Out Update to Tackle Network Congestion

A meme coin trading frenzy and a rapid spike in users has stressed the network in recent months.

Google files lawsuit against crypto scammers allegedly defrauding thousands

Tech giant Google has taken legal action against a group of crypto scammers, alleging their involvement in defrauding more than 100,000 people worldwide. Filed on Thursday, the lawsuit accuses the scammers of uploading fraudulent investment and cryptocurrency exchange apps to Google Play, Google’s official app store for Android devices.

Stacks Creator Ali Calls Bitcoin 'Apex Predator' as Development Flourishes on OG Blockchain

Muneeb Ali, the co-creator of Stacks and Princeton-educated computer scientist who's now CEO of the Bitcoin-focused development firm Trust Machines, spoke with CoinDesk's Jenn Sanasie on the flurry of development and layer-2 building now taking place on the original blockchain.

Understanding Bitcoin Halving: A beginner’s guide

The official blog for Trezor, the world’s first hardware wallet. All you need to keep your digital assets safe. - Medium

MIT taught a neural network how to show its work

2018-9-13 00:44

Santiment Network Token (SAN) на Currencies.ru

network neural → Результатов: 7

This neural network lets you build and evolve little AI creatures

2018-9-7 21:22

MIT’s depression-detecting AI might be its scariest creation yet

2018-9-6 21:44

IBM came up with a watermark for neural networks

2018-7-20 18:52

Experts warn DeepFakes could influence 2020 US election

2018-7-13 01:26

Facebook is teaching AI to talk you through directions

2018-7-11 19:00

AI Opinion ICO

2018-6-29 12:07

Партнеры Currencies.ru

Реклама на Currencies.ru