ResNet (2015): residual connections to go deeper
He et al. introduced residual networks that enabled training hundreds of layers and won ILSVRC 2015.
Sorted by date · America/Guayaquil timezone
Latest signals
Page 35 / 38 · 447 articles
GANs (2014): Goodfellow and the generator–discriminator game
Ian Goodfellow et al. introduced generative adversarial networks, opening an era of realistic image synthesis.
Word2Vec (2013): the algebra of words
Tomas Mikolov and Google colleagues released Word2Vec, popularizing dense embeddings and king–man+woman≈queen analogies.
AlexNet (2012): deep learning’s big bang in vision
Krizhevsky, Sutskever, and Hinton won ILSVRC 2012 with a deep CNN on GPUs, crashing classification error.
Deep pretraining before AlexNet: the table is set
Before 2012, Hinton, Bengio, and LeCun pushed greedy pretraining, autoencoders, and evidence that deep nets could learn.
From HMMs to DNNs in speech recognition
In the early 2010s, deep nets replaced key parts of HMM-GMM systems and sharply cut speech error rates.
IBM Watson wins Jeopardy! (2011): live QA before millions
Watson beat Jeopardy! champions by combining information retrieval, ranking, and many heuristics—not one magic net.
Netflix Prize (2006–2009): recommendations at competition scale
Netflix’s contest drove collaborative-filtering and ensemble research and popularized product ML.
ImageNet (2009): Fei-Fei Li and the dataset that changed vision
ImageNet offered millions of WordNet-organized labeled images, enabling the CNN breakthrough.
CUDA (2007): GPUs for machine learning
NVIDIA launched CUDA, enabling general-purpose GPU programming; without it, modern deep learning would have been impractical.
Boosting and random forests: ensembles that won competitions
AdaBoost, gradient boosting, and random forests dominated tabular ML and Kaggle contests before ubiquitous deep learning.
PageRank (1998): the web as a graph and the power of scale
Brin and Page’s PageRank showed how structural signals at internet scale predict relevance.