AI with Papers - Artificial Intelligence & Deep Learning
@AI_DeepLearning
15.2K
subscribers
135
photos
247
videos
14
files
1.31K
links
All the AI with papers. Every day fresh updates about
#DeepLearning
,
#MachineLearning
, LLMs and
#ComputerVision
Curated by Alessandro Ferrari |
https://www.linkedin.com/in/visionarynet/
#artificialintelligence
#machinelearning
#ml
#AI
Download Telegram
Join
AI with Papers - Artificial Intelligence & Deep Learning
15.2K subscribers
AI with Papers - Artificial Intelligence & Deep Learning
https://www.linkedin.com/posts/visionarynet_ai-artificialintelligence-machinelearning-activity-6855894398526480384-2-TD
Linkedin
Alessandro Ferrari on LinkedIn: #AI #AI #artificialintelligence
๐ฆ
๐ฆ
Non-Deep (โผ10 layers) Neural Networks by Intel Labs are super-competitive
๐ฆ
๐ฆ
๐
Deep Neural Networks (DNNs) have revolutionized the fields of #AI...
AI with Papers - Artificial Intelligence & Deep Learning
https://www.linkedin.com/posts/visionarynet_gpu-ai-artificialintelligence-activity-6856116669996183552-ouaP
Linkedin
Alessandro Ferrari on LinkedIn: #Nvidia #GPU #AI
๐
๐
โ
โ
THIS SNEAKER DOES NOT EXIST
โ
โ
๐
๐
Highlights:
โ
Thousands of fake shoes via neural generation
โ
Trained via #Nvidia StyleGAN2-ADA architecture
...
AI with Papers - Artificial Intelligence & Deep Learning
https://www.linkedin.com/posts/visionarynet_nvidia-selfdrivingcars-ai-activity-6856484092695908352-jUt0
Linkedin
Alessandro Ferrari on LinkedIn: #nvidia #selfdrivingcars #ai #artificialintelligence #machinelearningโฆ | 17 comments
๐ฝ
๐ฝ
#Nvidia just open-sourced DriveGAN, the most advanced neural simulator for #selfdrivingcars
๐ฝ
๐ฝ
๐
Realistic simulators are critical for training andโฆ | 17 comments on LinkedIn
AI with Papers - Artificial Intelligence & Deep Learning
https://www.linkedin.com/posts/visionarynet_ai-artificialintelligence-machinelearning-activity-6856934233420677120-wvdc
Linkedin
Alessandro Ferrari on LinkedIn: #ai #artificialintelligence #machinelearning #deeplearningaiโฆ
๐ฑโโ๏ธ
๐ฉโ๐ฆฐ
ไธๆตทไบค้ๅคงๅญฆ + Huawei open-sourcing CIPS-3D, a neural generator for 3D-aware synthetic faces
๐ฑโโ๏ธ
๐ฉโ๐ฆฐ
๐
StyleGAN achieved SOTA for generating HQโฆ
AI with Papers - Artificial Intelligence & Deep Learning
https://www.linkedin.com/posts/visionarynet_fashion-ai-artificialintelligence-activity-6856958259362971648-M3NS
Linkedin
Alessandro Ferrari on LinkedIn: #fashion #ai #ai #artificialintelligence #machinelearning #deeplearningaiโฆ
๐ฅ
๐ฝ
#Fashion gets high with fooling #AI patches
๐ฝ
๐ฅ
๐
UNLABELED develops adversarial patterns to protect people from labeling and beingโฆ
AI with Papers - Artificial Intelligence & Deep Learning
https://www.linkedin.com/posts/visionarynet_slam-ai-artificialintelligence-activity-6856988932891041792-9LWY
Linkedin
Alessandro Ferrari on LinkedIn: #slam #ai #artificialintelligence #machinelearning #deeplearningaiโฆ
๐ณ
๐ธ
University of Zurich open-sourced SVO Pro, one of the most solid framework for #slam and odometry
๐ธ
๐ณ
๐
SVO Pro features the support of different cameraโฆ
AI with Papers - Artificial Intelligence & Deep Learning
https://bit.ly/3C7t5R2
Linkedin
Alessandro Ferrari on LinkedIn: #ai #ai #artificialintelligence #machinelearning #deeplearningaiโฆ
๐ถ
๐ง
The ๐ค๐ฉ๐ช๐ญ๐ฅ-๐ฑ๐ข๐ณ๐ฆ๐ฏ๐ต paradigm in #AI to reach for the SOTA with simple fine-tuning and inversion
๐ถ
๐ง
๐
Transfer Learning refers to the process inโฆ
AI with Papers - Artificial Intelligence & Deep Learning
https://www.linkedin.com/posts/visionarynet_ai-artificialintelligence-machinelearning-activity-6857318823259971584-p1l7
Linkedin
๐ฅ
๐ฅ
CLOOB outperforms OpenAI's CLIP
๐ฅ
๐ฅ
๐
Contrastive Language-Image - Alessandro Ferrari on LinkedIn
๐ฅ
๐ฅ
CLOOB outperforms OpenAI's CLIP
๐ฅ
๐ฅ
๐
Contrastive Language-Image Pre-training (CLIP) yielded very impressive results at zero-shot transfer learning...
โค
1
AI with Papers - Artificial Intelligence & Deep Learning
https://www.linkedin.com/posts/visionarynet_ar-halloween-3d-activity-6858295079098990592-LLv1
Linkedin
Alessandro Ferrari on LinkedIn: #AR #halloween #3D
๐
๐
(Dystopic) #AR Human Duplicator: #halloween is coming
๐
๐
๐๐ข๐ ๐ก๐ฅ๐ข๐ ๐ก๐ญ๐ฌ:
โ
Snap's #3D body tracking and reconstruction
โ
"Hacking" a way to...
AI with Papers - Artificial Intelligence & Deep Learning
https://www.linkedin.com/posts/visionarynet_ai-artificialintelligence-machinelearning-activity-6858376126838964224-ce1U
Linkedin
Alessandro Ferrari on LinkedIn: #AI #artificialintelligence #machinelearning
๐คฏ
๐คฏ
Rutgers University open-sourced SLATE, a novel neural architecture for zero-shot image generation without text
๐คฏ
๐คฏ
๐
DALLยทE shown an impressive ability...
AI with Papers - Artificial Intelligence & Deep Learning
https://www.linkedin.com/posts/visionarynet_nvidia-ai-artificialintelligence-activity-6859130456873431040-OX_k
Linkedin
Alessandro Ferrari on LinkedIn: #Nvidia #AI #AI
๐ฝ
๐ฝ
#Nvidia open-sourced AugMax, a novel neural framework for data augmentation
๐ฝ
๐ฝ
๐
Out-of-distribution samples present a challenge when deploying #...
AI with Papers - Artificial Intelligence & Deep Learning
https://www.linkedin.com/posts/visionarynet_ai-artificialintelligence-machinelearning-activity-6859380973797212160-4MWN
Linkedin
Alessandro Ferrari on LinkedIn: #AI #AI #artificialintelligence | 17 comments
๐คฏ
๐คฏ
Facebook #AI open-sourced GHNs: hypernetwork to predict the parameters for new neural networks in <1 second!
๐คฏ
๐คฏ
๐
Is it possible to directly predict... 17 comments on LinkedIn
AI with Papers - Artificial Intelligence & Deep Learning
https://www.linkedin.com/posts/visionarynet_ai-artificialintelligence-machinelearning-activity-6861311579170897920-ZPPJ
Linkedin
Alessandro Ferrari on LinkedIn: #AI #artificialintelligence #machinelearning | 10 comments
โฐ
โฐ
German teams open-sourced Projected-GAN: training SOTA GANs 40x faster
โฐ
โฐ
๐
Generative Adversarial Networks produce high-quality images but are challenging... 10 comments on LinkedIn
AI with Papers - Artificial Intelligence & Deep Learning
https://www.linkedin.com/posts/visionarynet_valeo-selfdrivingcars-ai-activity-6861336996556673024-qzuJ
Linkedin
Alessandro Ferrari on LinkedIn: #Valeo #selfdrivingcars #AI
๐
๐
Detailed report by #Valeo on adversarial attacks in #selfdrivingcars
๐
๐
๐
Autonomous cars are expected to significantly reduce accidents. Despite...
AI with Papers - Artificial Intelligence & Deep Learning
https://www.linkedin.com/posts/visionarynet_ai-artificialintelligence-machinelearning-activity-6861352827613196288-tS0l
Linkedin
Alessandro Ferrari on LinkedIn: #AI #artificialintelligence #machinelearning
๐พ
๐พ
SenseTime ๅๆฑค็งๆ open-sourced GOF, a novel GRAF-like, 3D-surface aware, neural image generator
๐พ
๐พ
๐
Deep generative adversarial networks have demonstrated...
AI with Papers - Artificial Intelligence & Deep Learning
https://www.linkedin.com/posts/visionarynet_ai-artificialintelligence-machinelearning-activity-6861572759101734912-OhcG
Linkedin
Alessandro Ferrari on LinkedIn: #AI #artificialintelligence #machinelearning
๐
๐
CMU and UC Berkeley unveil a new policy to minimize energy consumption mixing walk, trot, and gallop
๐
๐
๐
Recent animal motor studies show that these...
AI with Papers - Artificial Intelligence & Deep Learning
https://www.linkedin.com/posts/visionarynet_ai-artificialintelligence-machinelearning-activity-6861647653512515584-uoO9
Linkedin
Alessandro Ferrari on LinkedIn: #ai #ai #artificialintelligence #machinelearning #deeplearningaiโฆ
๐คทโโ๏ธ
๐คทโโ๏ธ
Sberbank #AI just open-sourced the ๐๐ช๐จ๐จ๐๐๐ฃ DALLยทE
๐คทโโ๏ธ
๐คทโโ๏ธ
๐
The best, publicly available, AI-image generator takes text input in Russianโฆ
AI with Papers - Artificial Intelligence & Deep Learning
https://www.linkedin.com/posts/visionarynet_deepmind-imagenet-continuallearning-activity-6861981253357375489-y8Fm
Linkedin
Alessandro Ferrari on LinkedIn: #deepmind #imagenet #continuallearning #continuallearning #aiโฆ
๐น
๐น
Google #DeepMind unveils OPIN: training a model on #imagenet with #continuallearning strategy
๐น
๐น
๐
The novel One Pass ImageNet (OPIN) problem to studyโฆ
AI with Papers - Artificial Intelligence & Deep Learning
https://www.linkedin.com/posts/visionarynet_mit-ai-artificialintelligence-activity-6862293691676061696-j0HE
Linkedin
Alessandro Ferrari on LinkedIn: #mit #ai #artificialintelligence #machinelearning #deeplearningaiโฆ
๐ค
๐ค
#MIT shows impressive results on in-hand object re-orientation via imitation learning techniques
๐ค
๐ค
๐
In-hand object reorientation has been a challengingโฆ
AI with Papers - Artificial Intelligence & Deep Learning
https://www.linkedin.com/posts/visionarynet_ai-machinelearning-artificialintelligence-activity-6862433113939734528-2PXa