Just links
6.58K subscribers
358 photos
39 videos
10 files
7.76K links
That's just link aggregator of everything I consider interesting, especially DL and topological condensed matter physics. @EvgeniyZh
Download Telegram
Forwarded from эйай ньюз
NeuralRecon: Real-Time Coherent 3D Reconstruction from Monocular Video

Дождались. Исследователи разработали метод для риал-тайм 3D реконструкции сцены с помощью айфона.

Под капотом рекуррентная сетка, которая во время сканирования фьюзит TSDF представление со всех кадров, учитывая информацию с новых кадров онлайн.

И дрифт камеры в ARKit как-то решили, и есть код!

Это прям круто!

>> Сайт проекта >> Код
👍2
HyperMixer: An MLP-based Green AI Alternative to Transformers https://arxiv.org/abs/2203.03691
#mlp
Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing inference time https://arxiv.org/abs/2203.05482
90.94% top-1 ImageNet🤔
#cv #averaging
🤔7
The entropy of bulk quantum fields and the entanglement wedge of an evaporating black hole https://arxiv.org/abs/1905.08762
#physics #quantum_gravity #entropy
The Machine Learning for Combinatorial Optimization Competition (ML4CO): Results and Insights https://arxiv.org/abs/2203.02433
#competition #combinatorics #milp
Algebra of local symmetric operators and braided fusion n-category -- symmetry is a shadow of topological order https://arxiv.org/abs/2203.03596
#physics #symmetry #category
Quantum Many-Body Scars and Hilbert Space Fragmentation: A Review of Exact Results https://arxiv.org/abs/2109.00548
#physics #quantum #chaos #scars
The Role of ImageNet Classes in Fréchet Inception Distance
https://arxiv.org/abs/2203.06026
#cv #metrics #generative
Protocol to identify a topological superconducting phase in a three-terminal device https://arxiv.org/abs/2103.12217
#physics #majoranas #topological
Efficient Language Modeling with Sparse all-MLP https://arxiv.org/abs/2203.06850
#nlp #mlp
The Principle of Diversity: Training Stronger Vision Transformers Calls for Reducing All Levels of Redundancy https://arxiv.org/abs/2203.06345
#cv #vit
CLIP Models are Few-shot Learners: Empirical Studies on VQA and Visual Entailment https://arxiv.org/abs/2203.07190
#multimodal #zero_shot
Parametric exploration of zero-energy modes in three-terminal InSb-Al nanowire devices https://arxiv.org/abs/2203.00773
#physics #majoranas #topological
Forwarded from Artificial Intelligence
Bamboo: Building Mega-Scale Vision Dataset Continually with Human-Machine Synergy

Github: https://github.com/davidzhangyuanhan/bamboo

Project: https://opengvlab.shlab.org.cn/bamboo/home

Paper: https://arxiv.org/abs/2203.07845

@ArtificialIntelligencedl
👍1
Data-Driven Offline Optimization For Architecting Hardware Accelerators https://arxiv.org/abs/2110.11346
#rl #chips