#convnet
Why are ConvNets universal for translation-equivariant maps? Approximate one coordinate by an MLP, then share it across space. An accumulator trades width for depth. https://www.gpeyre.com/blog/2026/10/29/mlp-to-convnet-universality/
October 29, 2026 at 7:42 AM
Nah, most of those were invented in the "convnet on Imagenet" era.
October 2, 2026 at 5:06 PM
ConvNET is a forum where MLEs go to talk about their furry convention plans
September 23, 2026 at 4:29 PM
there’s no wonder yann lecun (the respected inventor of the ConvNet and ex-chief scientist of Meta) wonder whether the hugging face incident was staged,
September 20, 2026 at 12:20 AM
the punchline is we answered the mara question first, by accident: train him on engagement metrics and he tempts everyone in reach. meanwhile the convnet just sits there classifying, blameless, already liberated
September 14, 2026 at 3:24 AM
This makes me think of a conversation years ago where someone asked me 'how do we make a bodhisatva AI with no-self?' and I said something on the lines of 'that's easy, just train a convnet to classify stuff; what you should really be asking is, how to make a Mara-ist AI?'
September 14, 2026 at 3:22 AM
bender is not a computer scientist, gebru's background is in convnets, and i think meg's also a convnet person? all of them are perfectly well-qualified to comment on the subject but none of them are inventors of "the algorithm." (i would not describe a transformer as an algorithm.)
September 8, 2026 at 4:29 AM
In too much detail:

2012: comp. Physics grad school. Get really into gpu compute
2014: convolving halo densities with had-chosen filters / wavelets. What if we “fit” the filters? Find info on CNNs. A long hobby rabbit hole
2016-17: 3d convnet HOD idea falls apart when SPH cosmo sims discredited
I think mine would be more like:

Long time ago - before ChatGPT released:
One day AI will take over but who knows if it happens in my lifetime.

ChatGPT released:
Holy shit, it's happening!

Ever since:
There are no walls, superintelligence is just a matter of continued iteration.
September 3, 2026 at 2:29 AM
UniRepLKNet: A Universal Perception Large-Kernel ConvNet for Audio, Video,Point Cloud, Time-Series and Image Recognition arxiv.org/abs/2311.15599
December 24, 2023 at 4:40 AM
Es geht um die Objekterkennung in Drohnen aus Luftbildern mittels Convolutional Neural Networks (CNNs). Die Kameras erfassen große Bildbereiche, in denen Objekte oft sehr klein erscheinen.

Zwar von 2028, aber immer noch relevant.

arxiv.org/abs/1911.06073
Efficient ConvNet-based Object Detection for Unmanned Aerial Vehicles by Selective Tile Processing
Many applications utilizing Unmanned Aerial Vehicles (UAVs) require the use of computer vision algorithms to analyze the information captured from their on-board camera. Recent advances in deep learni...
arxiv.org
February 27, 2026 at 10:07 PM
2601.15621
本報告書では、高度な多言語対応、制御可能、堅牢性、ストリーミング対応を備えたテキスト音声変換モデル群であるQwen3-TTSシリーズを提示する。Qwen3-TTSは最先端の3秒音声クローン技術と記述ベース制御をサポートし、全く新しい音声の生成と出力音声の微細な操作を可能にします。10言語にまたがる500万時間...
January 26, 2026 at 12:07 AM
[24/30] 69 Likes, 1 Comments, 1 Posts
2311.15599, cs․CV | cs․AI | cs․LG, 27 Nov 2023

🆕UniRepLKNet: A Universal Perception Large-Kernel ConvNet for Audio, Video, Point Cloud, Time-Series and Image Recognition

Xiaohan Ding, Yiyuan Zhang, Yixiao Ge, Sijie Zhao, Lin Song, Xiangyu Yue, Ying Shan
December 13, 2023 at 12:05 AM
Look at its architecture:
1. The audio is cut into 80ms, which are sampled to 16 KHz (16000*0.08=1280 floats).
2. It is converted to a spectrogram,
3. A whisper-style encoder converts it to an audio token embedding through a convnet,
February 6, 2026 at 2:14 PM
I made a project to play dance dance revolution without a dance pad, instead using 2 high speed cameras, and running each frame through a shallow convnet to classify the steps.

The tape on the floor is so I know where to step. The stools are so I don't accidentally step on cameras
November 23, 2024 at 6:16 PM
I have a degree in cybernetics/robotics, been developing some AI since before convnet was the hotness, I have been using LLM's occasionally for both work and for fun since GPT2 because I'm interested in all types of AI technologically;
October 17, 2025 at 5:32 AM
Toufiq Musah, et al.: Brain Tumor Segmentation in Sub-Sahara Africa with Advanced Transformer and ConvNet Methods: Fine-Tuning, Data Mixing and Ensembling https://arxiv.org/abs/2508.10905 https://arxiv.org/pdf/2508.10905 https://arxiv.org/html/2508.10905
August 18, 2025 at 6:51 AM
> deepdream inceptionism resnet convnet activations, latent semantic interpolation, manifold loss landscape, visual vibrations
October 8, 2025 at 3:57 AM
> hallucinatory organic patterns, deepdream inceptionism, resnet convnet activations, latent semantic interpolation, visual vibrations, shifting colors, confusing perspective, traveling camera, shimmering iridescence
October 8, 2025 at 2:11 AM
"Yonas's Law" (after @ymt1234) You can *ALWAYS* rationalize why your #ConvNet has erroneously identified an object in an image!
September 21, 2016 at 6:27 PM
This looks to be a great set of posts on doing #DataScience on @Caltrain delays! Including #TensorFlow ConvNet! https://twitter.com/svdatascience/status/773230558616969216
September 7, 2016 at 3:05 AM
*GREAT* post about face recognition! It's not just "Run a ConvNet"; there is so much more! https://twitter.com/ageitgey/status/757241385602867200
July 25, 2016 at 4:24 PM
През 2012-та, AlexNet не е теоретичен пробив, а практически. Има оптимизирана имплементация, convnet написан на ниско ниво (през 2012-та, cuda е ниско ниво).

Междувременно, cuda надебеля. Сега тези имат оптимизирана имплементация на PTX ISA, което е едно ниво под cuda. За мен това е нещото.
January 30, 2025 at 7:18 AM
SPIDER: Spatial Image CorresponDence Estimator for Robust Calibration

Zhimin Shao, Abhay Yadav, Rama Chellappa, Cheng Peng

tl;dr: 3D VFM+2D ConvNet->feature extraction backbone; 3D descriptor head (for geometry)+2D warp head (for pattern) fusion

arxiv.org/abs/2511.17750
November 25, 2025 at 3:11 PM
Day 8: It felt like 7 hours and 42 experiments, but seems like fixing the sparsity loss term for my convnet was actually not that difficult. #100DaysOfCode
November 13, 2024 at 2:58 AM
Revisiting Contrastive Divergence for Density Estimation and Sample Generation

Azwar Abdulsalam, Joseph G. Makin

Action editor: Ying Nian Wu

https://openreview.net/forum?id=i5K4SZeqtq

#langevin #convnet #sampling
January 1, 2026 at 9:19 AM