L32: Momneum based gradient descent
Welcome to Lecture 32 of the course "Deep Learning" by Prof. Mitesh M.Khapra Full Course: https://study.iitm.ac.in/ds/course_pa... Video Overview This lecture introduces momentum based gradient descent an advanced optimization technique that improves learning efficiency by incorporating the history of past gradients. You will learn how momentum helps neural networks accelerate through regions of the loss surface with shallow slopes and reduces the chances of getting stuck in flat areas. The concept is explained using intuitive analogies followed by a breakdown of the underlying equations. We will compare momentum with standard gradient descent using visual examples to highlight how it changes the optimization trajectory. The lecture also discusses key behaviors such as overshooting and oscillations and raises important questions about how to minimize such effects in practice. By the end of this session you will have a clear understanding of the mechanics advantages and limitations of momentum and how it fits into the broader family of gradient based optimizers. About IIT Madras' online Bachelor of Science programme IIT Madras offers four-year BS programmes that aim to provide quality education to all, irrespective of age, educational background, or location. The BS programme has multiple levels, which provide flexibility to students to exit at any of these levels. Depending on the courses completed and credits earned, the learner can receive a Foundation Certificate from IITM CODE (Centre for Outreach and Digital Education), Diploma(s) from IIT Madras, or BSc/BS Degrees from IIT Madras. For more details, Visit: https://www.iitm.ac.in/academics/stud... #machinelearning #deeplearning #gradientdescent #momentum #optimization #algorithm #iitmadras #lectures #neuralnetworks #datascience #momentumoptimizer #gradienthistory #losssurface #trainingdynamics #mlalgorithms #backpropagation #momentumintuition #convergencespeed #deepnetworktraining #introtomomentum #overshooting #oscillations #adaptivelearning #gradientupdates #optimizationtechniques #neuraloptimization #mltrainingstrategy

L35: Nesterov accelarated gradient descent

Deep Learning(CS7015): Lec 5.4 Momentum based Gradient Descent

Activation Functions

Gradient Descent With Momentum (C2W2L06)

L42: AdaGrad: adaptive learning for sparse features

Why Aliens Would NEVER Invade Africa

L33: Gradient descent with adaptive learning rate in neural networks

From Child Prodigy to Winning Fields Medal, Nobel of Math

Vanishing & Exploding Gradient explained | A problem resulting from backpropagation

Gradient Descent with momentum and Steepest Descent

bounce + bounce = no bounce

Billionaire's WARNING: I'm SELLING. The Crash Is Already Here!

Why Do Predators Ignore Sleeping Humans?

Computing with Quantum bits Part-1

L34: Scheduling learning rate using decay & line search

You Know This Song (but the Orchestra Doesn’t) | Jacob Collier & VSO School of Music Orchestra | TED

Machine Learning Lecture 12 "Gradient Descent / Newton's Method" -Cornell CS4780 SP17

Momentum in SGD|Understanding Momentum in stochastic gradient descent

