KDnuggets

DECEMBER 16, 2019

It is an annual tradition for Xavier Amatriain to write a year-end retrospective of advances in AI/ML, and this year is no different. Gain an understanding of the important developments of the past year, as well as insights into what expect in 2020.

Artificial Intelligence 133

Artificial Intelligence Artificial Intelligence Machine Learning Machine Learning 133

Towards AI

JULY 23, 2023

Semi-Supervised Sequence Learning As we all know, supervised learning has a drawback, as it requires a huge labeled dataset to train. Generating Wikipedia By Summarizing Long Sequences This work was published by Peter J Liu at Google in 2019. But, the question is, how did all these concepts come together?

Natural Language Processing 98

Natural Language Processing Supervised Learning Deep Learning Deep Learning 98

Hacker News

MARCH 25, 2022

“Transformers made self-supervised learning possible, and AI jumped to warp speed,” said NVIDIA founder and CEO Jensen Huang in his keynote address this week at GTC. Transformers are in many cases replacing convolutional and recurrent neural networks (CNNs and RNNs), the most popular types of deep learning models just five years ago.

Machine Learning 182

Machine Learning Machine Learning Database AI 182

AssemblyAI

AUGUST 22, 2023

Using such data to train a model is called “supervised learning” On the other hand, pretraining requires no such human-labeled data. This process is called “self-supervised learning”, and is identical to supervised learning except for the fact that humans don’t have to create the labels.

Supervised Learning 97

Supervised Learning AI AI Machine Learning 97

Hacker News

FEBRUARY 14, 2023

It’s better than the top-word (zero temperature) case, but still at best a bit weird: This was done with the simplest GPT-2 model (from 2019). And many of the practical challenges around neural nets—and machine learning in general—center on acquiring or preparing the necessary training data. Here’s a random example.

Machine Learning 182

Machine Learning Machine Learning Algorithm Artificial Intelligence 182

DrivenData Labs

FEBRUARY 3, 2025

I love participating in various competitions involving deep learning, especially tasks involving natural language processing or LLMs. I generated unlabeled data for semi-supervised learning with Deberta-v3, then the Deberta-v3-large model was used to predict soft labels for the unlabeled data. Alejandro A.

Machine Learning 130

Machine Learning Machine Learning Data Science Natural Language Processing 130

Google Research AI blog

APRIL 27, 2023

In “ Inference of chronic obstructive pulmonary disease with deep learning on raw spirograms identifies new genetic loci and improves risk models ”, published in Nature Genetics , we’re excited to highlight a method for training accurate ML models for genetic discovery of diseases, even when using noisy and unreliable labels.

ML 99

ML ML Supervised Learning Deep Learning 99