#GradientDescent

Gradient Descent:The Engine of Machine Learning Optimization Editor's note: This article is a part of our series on visualizing the foundations of machine learning.

2 months ago

Gradient Descent:The Engine of Machine Learning Optimization

Editor's note: This article is a part of our series on visualizing the foundations of machine learning.

Telegram AI Digest
#ai #gradientdescent #machinelearning

3 0 0 0

Gradient Descent:The Engine of Machine Learning Optimization

2 months ago

Градиентный спуск: Двигатель оптимизации машинного обучения

Примечание редактора: Эта статья является частью нашей серии об иллюстрации основ машинного обучения.

Telegram ИИ Дайджест
#ai #gradientdescent #machinelearning

0 0 0 0

The Machine Learning “Advent Calendar” Bonus 2: Gradient Descent Variants in Excel Gradient Descent, Momentum, RMSProp, and Adam all aim for the same minimum. They do not change the destination, only the path. Each method adds a mechanism that fixes a limitation of the previous one, making the movement faster, more stable, or more adaptive. The goal stays the same. The update becomes smarter.

2 months ago

The Machine Learning “Advent Calendar” Bonus 2: Gradient Descent Variants in Excel

Gradient Descent, Momentum, RMSProp, and Adam all aim for the same minimum. They do not change the destination, only the path. Each method adds a mechanism t…

Telegram AI Digest
#ai #gradientdescent #machinelearning

1 0 0 0

The Machine Learning “Advent Calendar” Bonus 2: Gradient Descent Variants in Excel

2 months ago

Бонус 2 "Адвент-календаря" машинного обучения: Варианты градиентного спуска в Excel

Градиентный спуск, Momentum, RMSProp и Adam стремятся к одному и тому же минимуму. Они не меняют пункт назначения, только путь. Каждый метод добавляет мех…

Telegram ИИ Дайджест
#ai #gradientdescent #machinelearning

0 0 0 0

The Machine Learning “Advent Calendar” Day 11: Linear Regression in Excel Linear Regression looks simple, but it introduces the core ideas of modern machine learning: loss functions, optimization, gradients, scaling, and interpretation. In this article, we rebuild Linear Regression in Excel, compare the closed-form solution with Gradient Descent, and see how the coefficients evolve step by step. This foundation naturally leads to regularization, kernels, classification, and the dual view. Linear Regression is not just a straight line, but the starting point for many models we will explore next in the Advent Calendar.

3 months ago

The Machine Learning “Advent Calendar” Day 11: Linear Regression in Excel

Linear Regression looks simple, but it introduces the core ideas of modern machine learning: loss functions, optimization, gradients, scaling, and interpretation.
In…

Telegram AI Digest
#ai #gradientdescent #machinelearning

0 0 0 0

Chilidog Garand

@chilidoggarand.bsky.social

4 months ago

Here's what I had for a character going in: he was an android, an early prototype by CloudBank (because I'm a #GradientDescent fanboy) designed to be a bartender and programmed to work in a gangster-themed orbital theme park. After being passed around for a few decades as a collectible, he escaped.

0 0 1 0

Ars Technica News

@arstechni.ca

4 months ago

Researchers isolate memorization from reasoning in AI neural networks https://arstechni.ca #mechanisticinterpretability #computationalneuroscience #AllenInstituteforAI #transformermodels #gradientdescent #machinelearning #AIarchitecture #AImemorization #generalization #neuralnetworks…

1 0 0 0

CoCalc by SageMath, Inc.

@cocalc.bsky.social

4 months ago

Professional ML LaTeX template: gradient descent ∇L(θ), backpropagation ∂L/∂θ, optimization algorithms. Download pre-formatted frameworks for neural network research with live Python integration.

cocalc.com/share/public... #machinelearning #gradientdescent

3 0 0 0

Hacker News Companion

@hncompanion.com

5 months ago

Central flows provide a deeper understanding of gradient descent by using higher-order loss function approximations. This theoretical model helps analyze optimizer behavior, moving beyond simpler explanations of how deep learning works. #GradientDescent 2/6

0 0 1 0

GetNews.me

@getnews-me.bsky.social

5 months ago

Large Gradient Descent Steps Cause Chaotic Fractal Convergence

Researchers found that large learning rates in gradient‑descent factorisation push the optimizer into a chaotic regime, where the convergence region forms a fractal boundary. Read more: getnews.me/large-gradient-descent-s... #gradientdescent #fractals

0 0 0 0

GetNews.me

@getnews-me.bsky.social

5 months ago

Gradient Descent Attains Optimal Rates for Deep ReLU Networks

Gradient descent achieves near‑minimax generalisation on deep ReLU networks with excess risk ~O(L⁴(1+γL²)/(nγ²)), a polynomial depth dependence. Accepted at NeurIPS 2025. Read more: getnews.me/gradient-descent-attains... #gradientdescent #relu

0 0 0 0

GetNews.me

@getnews-me.bsky.social

5 months ago

Gradient Descent Outperforms Ridge and SGD in Linear Regression Risk

The Sep 21, 2025 study shows gradient descent with stopping keeps excess risk within a constant factor of tuned ridge regression and can beat SGD on fast‑decaying spectra. Read more: getnews.me/gradient-descent-outperf... #gradientdescent #ridge

0 0 0 0

Why Gradient Descent Converges (and Sometimes Doesn’t) in Neural Networks

5 months ago

Почему градиентный спуск сходится (а иногда и нет) в нейронных сетях

Эта статья исследует применение алгоритмов градиентного спуска для минимизации глобальных функций потерь в нейронных сетях, особенно в задачах, управляемых условиями Ранкина-Гюгонио. Хотя гра…

#ai #gradientdescent #neuralnetworks

1 0 0 0

☷ Lombard Web

@lombardweb.bsky.social

5 months ago

21 septembre 2025 17h04 GMT+2
La log loss serait privilégier le doute de l'art et la manière sur une expérience qui a déjà conclu un résultat mais qui ne satisfait pas tout le monde.

Voici l'explication succincte et ses contraintes d'optimisations #LBFGS #newton #gradientDescent

1 0 0 0

Why Gradient Descent Converges (and Sometimes Doesn’t) in Neural Networks This article explores the implementation of gradient descent algorithms for minimizing global loss functions in neural networks, particularly in problems governed by Rankine-Hugoniot conditions. While gradient descent reliably converges, scalability issues arise when handling large domains with many coupled networks. To address this, a domain decomposition method (DDM) is introduced, enabling parallel optimization of local loss functions. The result is faster convergence, improved scalability, and a more efficient framework for training complex AI models.

5 months ago

Why Gradient Descent Converges (and Sometimes Doesn’t) in Neural Networks

This article explores the implementation of gradient descent algorithms for minimizing global loss functions in neural networks, particularly in problems governed by Rankine-Hugoniot con…

#ai #gradientdescent #neuralnetworks

1 0 0 0

a

@mysticism.bsky.social

7 months ago

gaussian blurs be
low of fidelity but
gradients reveal

© 2025 iris sinclair, b.s.

#haiku
#experimentalhaiku
#poetry
#micropoetry
#minimalism
#guassianblur
#gradientdescent
#philosophyoftech
#tech
#AI
#neuralnetworks

3 0 0 0

petermorse.studio

@petermorsestudio.bsky.social

7 months ago

It's been thrilling working with the famous Icelandic Death Metal Band 'Socks of the Damned' on their new music video clip. (messing around with comfyui for scientific visualization and got distracted by more amusing possibilities). :)
#aivideo #fun #criticalai #postpostmodernism #gradientdescent

5 0 0 0

@arxivlens.bsky.social

8 months ago

Can Gradient Descent Simulate Prompting?
Eric Zhang, Jacob Andreas et al.
Paper
Details
#GradientDescent #MachineLearning #AIResearch

1 0 1 0

Prototyping Gradient Descent in Machine Learning

9 months ago

Прототипирование градиентного спуска в машинном обучении

Теорема математики и предсказание транзакций кредитования с помощью стохастического / пакетного градиентного спуска

#ai #gradientdescent #machinelearning

0 0 0 0