Artificial Intelligence
16.3K subscribers
1.09K photos
7 videos
1 file
1.95K links
Artificial Intelligence

admin - @haarrp

@itchannels_telegram - 🔥 best it channels

@ai_machinelearning_big_data - Machine learning channel

@pythonl - Our Python channel

@pythonlbooks- python книги📚

@datascienceiot - ml 📚

РКН: clck.ru/3FmwZw
Download Telegram
✔️ CITADEL: Conditional Token Interaction via Dynamic Lexical Routing for Efficient and Effective Multi-Vector Retrieval

🖥 Github: https://github.com/facebookresearch/dpr-scale

🗒 Paper: https://arxiv.org/abs/2211.10411v1

➡️ Dataset: https://paperswithcode.com/dataset/ms-marco

@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
👍51👎1🔥1🥰1
⭐️ Traffic4cast 2022 Competition: from few public vehicle counters to entire city-wide traffic

🖥 Github: https://github.com/iarai/neurips2022-traffic4cast

🗒 Paper: https://arxiv.org/abs/2211.09984v1

➡️ Dataset: https://developer.here.com/sample-data

@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
4👍2👎1🔥1
Please open Telegram to view this post
VIEW IN TELEGRAM
3👍2🥰1
Please open Telegram to view this post
VIEW IN TELEGRAM
5👍2🔥2
На международной онлайн-конфереции про искусственный интеллек AI Journey ученые и эксперты области обсуждают самые разные темы. Одна из них — применение нелинейных ядер.

Интересное мнение по этому поводу высказал Дмитрий Девяткин, научный сотрудник Института системного программирования РАН:

«Эксперименты показали, что применение нелинейных ядер позволяет на ряде наборов данных улучшить качество классификации. Используемый метод усиления и снижения сложности ансамбля деревьев решений показал свою применимость для улучшения ансамблей деревьев решений с линейными и нелинейными разделителями. Тем не менее было отмечено, что при построении нелинейных разделителей на больших наборах данных скорость всё-таки недостаточная для анализа больших данных, поэтому актуальной осталась задача разработки вычислительно эффективных методов двойственной оптимизации при условии масштабирования переменных невязки».
👍4
Please open Telegram to view this post
VIEW IN TELEGRAM
👍6🔥21
👣 Flow: Joint Semantic and Style Editing of Facial Images

git clone https://github.com/visinf/s2-flow.git
cd s2-flow/


🖥 Github: https://github.com/visinf/s2-flow

🗒 Paper: https://arxiv.org/abs/2211.12209

➡️ Dataset: https://drive.google.com/drive/folders/1nZ_U0qCFFwBM9L_h9mmV9W0axFt4Xd-N?usp=sharing


@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
4👍1🔥1
⭐️ Texts as Images in Prompt Tuning for Multi-Label Image Recognition

🖥 Github: https://github.com/guozix/tai-dpt

🗒 Paper: https://arxiv.org/abs/2211.12739v1

➡️ Dataset: https://paperswithcode.com/dataset/nus-wide

@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
👍31🔥1
🖥 EurNet: Efficient Multi-Range Relational Modeling of Spatial Multi-Relational Data


🖥 Github: https://github.com/hirl-team/eurnet-image

🗒 Paper: https://arxiv.org/abs/2211.12941v1

➡️ Dataset: https://paperswithcode.com/dataset/ade20k

@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
👍51🔥1
EHRSQL: A Practical Text-to-SQL Benchmark for Electronic Health Records

🖥 Github: https://github.com/glee4810/EHRSQL

🗒 Paper: https://openreview.net/forum?id=B2W8Vy0rarw

➡️ Dataset: https://paperswithcode.com/dataset/mimic-iii

@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
👍81🔥1
An Incremental Learning Approach for Sustainable Region Isolation and Integration

🖥 Github: https://github.com/Wuziyi123/SRII

🗒 Paper: https://openreview.net/forum?id=Hkf2_sC5FX

➡️ Dataset: https://paperswithcode.com/dataset/miniimagenet-1

@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
2👍2🥰1
Please open Telegram to view this post
VIEW IN TELEGRAM
Please open Telegram to view this post
VIEW IN TELEGRAM
👍52🔥2
🖥 MGTANet

A novel 3D object detection architecture, which can encode LiDAR point cloud sequences acquired by multiple successive scans.

🖥 Github: https://github.com/hyjhkoh/mgtanet

📌 Project: https://sites.google.com/view/junhokoh/aaai2023?authuser=0

🗒 Paper: https://arxiv.org/abs/2212.00442v1

➡️ Dataset: https://paperswithcode.com/dataset/nuscenes

@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
👍6
💨 Improving Zero-Shot Models with Label Distribution Priors

🖥 Github: https://github.com/jonkahana/clippr

🗒 Paper: https://arxiv.org/abs/2212.00784v1

➡️ Dataset: https://paperswithcode.com/dataset/stanford-cars

@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
👍4
Please open Telegram to view this post
VIEW IN TELEGRAM
👍52🥰1
⭐️ RLogist: Fast Observation Strategy on Whole-slide Images with Deep Reinforcement Learning


🖥 Github: https://github.com/tencent-ailab/rlogist

🗒 Paper: https://arxiv.org/abs/2212.01737v1

➡️ Dataset: https://paperswithcode.com/dataset/nus-wide

@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
👍42🔥1