GotAI.NET

Форум: Проблемы искусственного интеллекта

 

Регистрация | Вход

 Все темы | Новая тема Стр.1 (1)   Поиск:  
 Автор Тема: Waymo
Ilya Geller
Сообщений: 4882
Waymo
Добавлено: 07 апр 20 3:01
https://www.intelligenttransport.com/transport-news/97860/waymo-and-google-brain-partner-to-advance-data-augmentation-research/
In order to advance its machine learning models and further improve its self-driving system’s perception, Waymo, the self-driving car service, has teamed up with colleagues from Google Brain, a deep learning artificial intelligence research team at Google, to extend its automated data augmentation research and test it against its dataset of autonomous driving.

Пиздят суки, они изначально основали Waymo на моем текстуальном поиске.
[Ответ][Цитата]
Ilya Geller
Сообщений: 4882
На: Waymo
Добавлено: 15 апр 20 11:02
Users can upload pictures on the official website www.captionbot.ai. After a while, they can see the description of pictures in the system. Although its accuracy is not low, it still needs to be improved. Taking a once popular black question mark facial expression picture as an example, AI quickly gave an objective answer: "I think it's basketball player Nick Young showing his teeth and smiling."
http://www.digitaljournal.com/pr/4651506#ixzz6JhJuxKRH

Это технология используется Хуйглем для управления автомобилями.
[Ответ][Цитата]
Ilya Geller
Сообщений: 4882
На: Waymo
Добавлено: 16 апр 20 16:11
STARTUPS Waymo’s Self-Driving Technology Gets Smarter, Recognizes Billions of Objects Thanks To Content Search
To solve this problem, as VentureBeat reports, Waymo recently developed a tool dubbed “Content Search”, which functions similarly to how Google Image Search and Google Photos operate. These systems match queries with the semantic content within images, generating representations of the objects that make image retrieval based on natural language queries easier.
https://www.unite.ai/waymos-self-driving-technology-gets-smarter-thanks-to-content-search/

Ну идите, идите ко мне пидорки гнойные!
[Ответ][Цитата]
Ilya Geller
Сообщений: 4882
На: Waymo
Добавлено: 08 июн 20 11:14
https://www.unite.ai/waymos-self-driving-technology-gets-smarter-thanks-to-content-search/

Waymo recently developed a tool dubbed “Content Search”, which functions similarly to how Google Image Search and Google Photos operate. These systems match queries with the semantic content within images, generating representations of the objects that make image retrieval based on natural language queries easier.

[Ответ][Цитата]
Ilya Geller
Сообщений: 4882
На: Waymo
Добавлено: 16 июн 20 12:17
Судя по всему дела у Waymo плохи:

Waymo, the self-driving arm of Google’s parent company Alphabet, has raised $2.25bn, the first time the venture outside money since the project began development more than 11 years ago.

11 лет садят миллиарды и ничего нет.
[Ответ][Цитата]
Ilya Geller
Сообщений: 4882
На: Waymo
Добавлено: 22 июн 20 12:35
Так вот на что ушли миллиарды!

Unlabelled data is learned in a task-agnostic way in the pretraining phase, which means the model has no prior classification knowledge. The researchers find that using a deep and wide neural network can be more label-efficient and greatly improve accuracy.
Network size is the key at this and the following phase. Supervised labels are used in the fine-tuning stage to further refine accuracy. Here the team found that using fewer labelled examples helps the bigger and deeper network improve accuracy. The researchers also discovered that the task-specific prediction can be further distilled to a smaller network simply by labelling the unlabelled data again.
https://syncedreview.com/2020/06/22/google-brains-simclrv2-achieves-new-sota-in-semi-supervised-learning/

Царьгородцеву каюк.
[Ответ][Цитата]
Ilya Geller
Сообщений: 4882
На: Waymo
Добавлено: 22 июн 20 12:40
Цитата:
Автор: Ilya Geller

Так вот на что ушли миллиарды!

Unlabelled data is learned in a task-agnostic way in the pretraining phase, which means the model has no prior classification knowledge. The researchers find that using a deep and wide neural network can be more label-efficient and greatly improve accuracy.
Network size is the key at this and the following phase. Supervised labels are used in the fine-tuning stage to further refine accuracy. Here the team found that using fewer labelled examples helps the bigger and deeper network improve accuracy. The researchers also discovered that the task-specific prediction can be further distilled to a smaller network simply by labelling the unlabelled data again.
https://syncedreview.com/2020/06/22/google-brains-simclrv2-achieves-new-sota-in-semi-supervised-learning/

Царьгородцеву каюк.


The updated framework takes the “unsupervised pretrain, supervised fine-tune” paradigm popular in natural language processing and applies it to image recognition. Unlabelled data is learned in a task-agnostic way in the pretraining phase, which means the model has no prior classification knowledge

[Ответ][Цитата]
 Стр.1 (1)