Lmst

'The Effect of SGD Batch Size on Autoencoder Learning: Sparsity, Sharpness, and Feature Learning', by Nikhil Ghosh, Spencer Frei, Wooseok Ha, Bin Yu.

http://jmlr.org/papers/v26/23-1022.html

#sgd #autoencoder #sparse

UEBA в кибербезе: как профилирование поведения пользователей на основе Autoencoder помогает выявлять угрозы и аномалии

В современном мире количество атак растёт пропорционально количеству внедрений новых технологий, особенно когда технологии ещё недостаточно изучены. В последнее время атаки становятся всё более разнообразными, а методы их реализации — всё более изощрёнными. Дополнительные проблемы несут и методы искусственного интеллекта, которыми вооружаются специалисты RedTeam. В руках опытного специалиста эти инструменты становятся реальной угрозой безопасности потенциальных целей. Большинство средств информационной безопасности основаны на корреляционных или статистических методах, которые в современных реалиях часто оказываются неэффективными. Что же тогда остаётся специалистам BlueTeam?

https://habr.com/ru/companies/gaz-is/articles/899058/

#газинформсервис #информационная_безопасность #ueba #поведенческая_аналитика #lstm #autoencoder #falco

SciTech Chronicles. . . . . . . . .Mar 28th, 2025

https://bit.ly/stc032825

#Prototaxites #Devonian #Rhynie #chitin #cellulose #delignifying #lye #diethylenetriamine #ACC #hyperglycemic #hippocampus #anhedonia #ANNs #fMRI #autoencoder #RSA #dairy #high-value #reuse #upcycle

Автоэнкодеры простыми словами

Автоэнкодеры являются базовой техникой машинного обучения и искусственного интеллекта, на основе которой строятся более сложные модели, например, в диффузионных моделях, таких как Stable Diffusion. Что же такое автоэнкодер?

https://habr.com/ru/companies/raft/articles/851548/

#autoencoder #автоэнкодер #машинное+обучение

'Representation Learning via Manifold Flattening and Reconstruction', by Michael Psenka, Druv Pai, Vishal Raman, Shankar Sastry, Yi Ma.

http://jmlr.org/papers/v25/23-0615.html

#flatnet #autoencoder #manifold

These four #Python #tutorials introduce and discuss #PCA, #tsne, #factoranalysis, and #Autoencoder as powerful tools for #DimensionalityReduction:

🌍 https://www.fabriziomusacchio.com/blog/2023-06-16-pca_with_python/
🌍 https://www.fabriziomusacchio.com/blog/2023-06-12-tsne_vs_pca/
🌍 https://www.fabriziomusacchio.com/blog/2023-06-16-factoranalysis_with_python/
🌍 https://www.fabriziomusacchio.com/blog/2023-06-16-autoencoder_with_python/

Feel free to share, use and remix 😊🙏

#tutorial

Here is a #Python #tutorial, where I compare #PCA, #KernelPCA (#kPCA), #Autoencoder and #CNN with regard to #ImageDenoising 🌁 🧹

🌍 https://www.fabriziomusacchio.com/blog/2023-06-20-ai_image_denoising/

Compression of Structured Data with Autoencoders: Provable Benefit of Nonlinearities and Depth
Kevin Kögler, Alexander Shevchenko, Hamed Hassani, Marco Mondelli

abs: http://arxiv.org/abs/2402.05013
pdf: https://arxiv.org/pdf/2402.05013.pdf

#arXiv #ComputerVision #Autoencoder

Autoencoders are a prominent model in many empirical branches of machine
learning and lossy data compression. However, basic theoretical questions
remain unanswered even in a shallow two-layer setting. In particular, to what
degree does a shallow autoencoder capture the structure of the underlying data
distribution? For the prototypical case of the 1-bit compression of sparse
Gaussian data, we prove that gradient descent converges to a solution that
completely disregards the sparse structure of the input. Namely, the
performance of the algorithm is the same as if it was compressing a Gaussian
source - with no sparsity. For general data distributions, we give evidence of
a phase transition phenomenon in the shape of the gradient descent minimizer,
as a function of the data sparsity: below the critical sparsity level, the
minimizer is a rotation taken uniformly at random (just like in the compression
of non-sparse data); above the critical sparsity, the minimizer is the identity
(up to a permutation). Finally, by exploiting a connection with approximate
message passing algorithms, we show how to improve upon Gaussian performance
for the compression of sparse data: adding a denoising function to a shallow
architecture already reduces the loss provably, and a suitable multi-layer
decoder leads to a further improvement. We validate our findings on image
datasets, such as CIFAR-10 and MNIST.

Os VAE são fascinantes...😮

https://youtu.be/pea3sH6orMc?si=YXY5U3I5WBn6HcPu

#autoencoder #variational #diffusion #VAE

A deep variational #autoencoder for #proteomics #mass_spectrometry #data analysis.

https://phys.org/news/2023-11-deep-variational-autoencoder-proteomics-mass.html

Würstchen - Schnelle Diffusion jetzt für die Bildgenerierung

#KI #AI #Würstchen #Diffusionsmodell #Bildgenerierung #VQGAN #Autoencoder #GPU #Effizienz #PyTorch #FlashAttention #TorchCompile #kuenstlicheintelligenz #künstlicheintelligenz

https://kinews24.de/wuerstchen-schnelle-diffusion

A simple, efficient and scalable contrastive masked autoencoder for learning visual representations

https://openreview.net/forum?id=pjdxPts6er

#autoencoders #autoencoder #imagenet

Extremely proud to present my group's pre-print paper executed by my PhD student Mr. Jonas Köhler on #earthquake #forecasting using #deeplearning. We pose earthquake forecasting as a classification problem and train a Neural Network to decide, whether a #timeseries of length greater than 2 years will end in an earthquake on the following day with magnitude greater than 5 or not. The study could be summarised in following three points: 1) We use spatio-temporal b value data, on which we train an #autoencoder to learn the normal seismic behaviour. 2) We then take the pixel by pixel reconstruction error as input for a Convolutional Dilated Network classifier, whose model output could serve for earthquake forecasting. 3) We develop a special progressive training method for this model to mimic real life use. #data #research #ai4good #ai4science #seismology
https://arxiv.org/abs/2307.01812

A Latent Diffusion Model for Protein Structure Generation

https://openreview.net/forum?id=8zzjemksJd

#proteins #autoencoder #protein

CAE v2: Context Autoencoder with CLIP Latent Alignment

https://openreview.net/forum?id=f36LaK7M0F

#autoencoder #encoder #masked

"Autoencoder Image Interpolation by Shaping the Latent Space"

🔗: https://arxiv.org/abs/2008.01487# #arxiv #interpolation #autoencoder

"Diffusion language models" blog post by @sedielem - lots of illuminating details, e.g. the connection between diffusion noise levels and scale (frequency) of features https://benanne.github.io/2023/01/09/diffusion-language.html #AI #deeplearning #diffusion #autoencoder