1663811049_maxresdefault.jpg
39
12:51

Early Stopping in PyTorch to Prevent Overfitting (3.4)

It can be difficult to know how many epochs to train a neural network for. Early stopping stops the neural network from training before it begins to seriously overfitting. Generally too many epochs will result in an overfit neural network and too few will be underfit. Code for This Video: https://github.com/jeffheaton/t81_558_deep_learning/blob/pytorch/t81_558_class_03_4_early_stop.ipynb ~~~~~~~~~~~~~~~ COURSE MATERIAL ~~~~~~~~~~~~~~~ […]
1658796414_maxresdefault.jpg
725
01:36:58

Learn PyTorch for deep learning in a day. Literally.

Welcome to the most beginner-friendly place on the internet to learn PyTorch for deep learning. All code on GitHub – https://dbourke.link/pt-github Ask a question – https://dbourke.link/pt-github-discussions Read the course materials online – https://learnpytorch.io Sign up for the full course on Zero to Mastery (20+ hours more video) – https://dbourke.link/ZTMPyTorch Below are the timestamps/outline of the […]
1645816656_maxresdefault.jpg
96
51:48

#64 Prof. Gary Marcus 3.0 [Unplugged]

Patreon: https://www.patreon.com/mlst Discord: https://discord.gg/HNnAwSduud We have a chat with Prof. Gary Marcus about everything which is currently top of mind for him, consciousness [00:00:00] Gary intro [00:01:25] Slightly conscious [00:24:59] Abstract, compositional models [00:32:46] Spline theory of NNs [00:36:17] Self driving cars / algebraic reasoning [00:39:43] Extrapolation [00:44:15] Scaling laws [00:49:50] Maximum likelihood estimation Pod […]
1640391879_maxresdefault.jpg
240
03:34

Two Minute Papers: Deep Learning and Cancer Research | Two Minute Papers #64

A few quite exciting applications of deep learning in cancer research have appeared recently. This new algorithm can recognize cancer cells by looking at blood samples without introducing any intrusive chemicals in the process. Amazing results ahead. 🙂 _________________________ The paper “Deep Learning in Label-free Cell Classification” is available here: http://www.nature.com/articles/srep21471 The link from Healthline: […]
1640370184_hqdefault.jpg
285
03:29

Two Minute Papers: Training Deep Neural Networks With Dropout | Two Minute Papers #62

In this episode, we discuss the bane of many machine learning algorithms – overfitting. It is also explained why it is an undesirable way to learn and how to combat it via dropout. _____________________ The paper “Dropout: A Simple Way to Prevent Neural Networks from Overtting” is available here: https://www.cs.toronto.edu/~hinton/absps/JMLRdropout.pdf Andrej Karpathy’s autoencoder is available […]
1640348415_maxresdefault.jpg
304
04:34

Two Minute Papers: Overfitting and Regularization For Deep Learning | Two Minute Papers #56

In this episode, we discuss the bane of many machine learning algorithms – overfitting. It is also explained why it is an undesirable way to learn and how to combat it via L1 and L2 regularization. _____________________________ The paper “Regression Shrinkage and Selection via the Lasso” is available here: http://statweb.stanford.edu/~tibs/lasso/lasso.pdf Andrej Karpathy’s excellent lecture notes […]
1639221123_hqdefault.jpg
863
04:46

Two Minute Papers: No Such Thing As Artificial Intelligence | Two Minute Papers #60

What is happening with the neural networks in this video? You’ll find the answers in these videos: 1. How Does Deep Learning Work? – https://www.youtube.com/watch?v=He4t7Zekob0&index=5&list=PLujxSBD-JXglGL3ERdDOhthD3jTlfudC2 2. Overfitting and Regularization For Deep Learning – https://www.youtube.com/watch?v=6aF9sJrzxaM&index=18&list=PLujxSBD-JXglGL3ERdDOhthD3jTlfudC2 In this episode, we discuss the perils of debating whether different existing techniques can be deemed artificial intelligence or not. ____________________ […]
1638064870_maxresdefault.jpg
3
04:55

CoRL 2020, Spotlight Talk 108: Learning hierarchical relationships for object-goal navigation

“**Learning hierarchical relationships for object-goal navigation** Anwesan Pal (UC San Diego); Yiding Qiu (UC San Diego)*; Henrik Christensen (UC San Diego) Publication: http://corlconf.github.io/paper_108/ **Abstract** Direct search for objects as part of navigation poses a challenge for small items. Utilizing context in the form of object-object relationships enable hierarchical search for targets efficiently. Most of the […]
1637071207_maxresdefault.jpg
2.14K
04:32

Two Minute Papers: This Neural Network Optimizes Itself | Two Minute Papers #212

The paper “Hierarchical Representations for Efficient Architecture Search” is available here: https://arxiv.org/pdf/1711.00436.pdf Genetic algorithm (+ Mona Lisa problem) implementation: 1. https://users.cg.tuwien.ac.at/zsolnai/gfx/mona_lisa_parallel_genetic_algorithm/ 2. https://users.cg.tuwien.ac.at/zsolnai/gfx/knapsack_genetic/ Andrej Karpathy’s online demo: http://cs.stanford.edu/people/karpathy/convnetjs/demo/classify2d.html Overfitting and Regularization For Deep Learning – https://www.youtube.com/watch?v=6aF9sJrzxaM Training Deep Neural Networks With Dropout – https://www.youtube.com/watch?v=LhhEv1dMpKE How Do Genetic Algorithms Work? – https://www.youtube.com/watch?v=ziMHaGQJuSI We would like to […]
1636164313_maxresdefault.jpg
4.02K
05:53

Neural Networks Demystified [Part 7: Overfitting, Testing, and Regularization]

We’ve built and trained our neural network, but before we celebrate, we must be sure that our model is representative of the real world. Supporting Code: https://github.com/stephencwelch/Neural-Networks-Demystified Nate Silver’s Book: http://www.amazon.com/Signal-Noise-Many-Predictions-Fail/dp/159420411X/ref=sr_1_1?ie=UTF8&qid=1421442340&sr=8-1&keywords=signal+and+the+noise Caltech Machine Learning Course: https://work.caltech.edu/telecourse.html And the lecture shown: http://youtu.be/Dc0sr0kdBVI?t=56m52s In this series, we will build and train a complete Artificial Neural Network in […]
1636160644_maxresdefault.jpg
0

Neural Networks Demystified [Part 6: Training]

After all that work it’s finally time to train our Neural Network. We’ll use the BFGS numerical optimization algorithm and have a look at the results. Supporting Code: https://github.com/stephencwelch/Neural-Networks-Demystified Yann Lecun’s Efficient BackProp Paper: http://yann.lecun.com/exdb/publis/pdf/lecun-98b.pdf More on BFGS: http://en.wikipedia.org/wiki/Broyden%E2%80%93Fletcher%E2%80%93Goldfarb%E2%80%93Shanno_algorithm In this series, we will build and train a complete Artificial Neural Network in python. New […]
1636156587_maxresdefault.jpg
0

Neural Networks Demystified [Part 5: Numerical Gradient Checking]

When building complex systems like neural networks, checking portions of your work can save hours of headache. Here we’ll check our gradient computations. Supporting code: https://github.com/stephencwelch/Neural-Networks-Demystified Link to excellent Stanford tutorial: http://ufldl.stanford.edu/wiki/index.php/UFLDL_Tutorial In this series, we will build and train a complete Artificial Neural Network in python. New videos every other friday. Part 1: Data […]
1636152945_maxresdefault.jpg
0

Neural Networks Demystified [Part 4: Backpropagation]

Backpropagation as simple as possible, but no simpler. Perhaps the most misunderstood part of neural networks, Backpropagation of errors is the key step that allows ANNs to learn. In this video, I give the derivation and thought processes behind backpropagation using high school level calculus. Supporting Code and Equations: https://github.com/stephencwelch/Neural-Networks-Demystified In this series, we will […]
1636149267_maxresdefault.jpg
5.01K
06:56

Neural Networks Demystified [Part 3: Gradient Descent]

Neural Networks Demystified @stephencwelch Supporting Code: https://github.com/stephencwelch/Neural-Networks-Demystified Link to Yann’s Talk: In this short series, we will build and train a complete Artificial Neural Network in python. New videos every other friday. Part 1: Data + Architecture Part 2: Forward Propagation Part 3: Gradient Descent Part 4: Backpropagation Part 5: Numerical Gradient Checking Part 6: […]
1636145619_maxresdefault.jpg
5.27K
04:28

Neural Networks Demystified [Part 2: Forward Propagation]

Neural Networks Demystified @stephencwelch Supporting Code: https://github.com/stephencwelch/Neural-Networks-Demystified In this short series, we will build and train a complete Artificial Neural Network in python. New videos every other friday. Part 1: Data + Architecture Part 2: Forward Propagation Part 3: Gradient Descent Part 4: Backpropagation Part 5: Numerical Gradient Checking Part 6: Training Part 7: Overfitting, […]
1636141873_maxresdefault.jpg
0

Neural Networks Demystified [Part 1: Data and Architecture]

Neural Networks Demystified Part 1: Data and Architecture @stephencwelch Supporting Code: https://github.com/stephencwelch/Neural-Networks-Demystified In this short series, we will build and train a complete Artificial Neural Network in python. New videos every other friday. Part 1: Data + Architecture Part 2: Forward Propagation Part 3: Gradient Descent Part 4: Backpropagation Part 5: Numerical Gradient Checking Part […]
1635765186_maxresdefault.jpg
30

Data Science in 4 Minutes: Quick High Level Overview

In this video I discuss classification, regression, overfitting, underfitting, bias, variance, and feature engineering. This gives a very quick overview of data science in 4 minutes. Source of this machine learning/AI Video
1635725869_maxresdefault.jpg
5

Buy AND Build for Production Machine Learning with Nir Bar-Lev – #488

Today we’re joined by Nir Bar-Lev, co-founder and CEO of ClearML. In our conversation with Nir, we explore how his view of the wide vs deep machine learning platforms paradox has changed and evolved over time, how companies should think about building vs buying and integration, and his thoughts on why experiment management has become […]
1635646500_maxresdefault.jpg
47
13:20

Top Kaggle Solution for Spring 2020 Kaggle In-Class Paperclips Challenge

In this video, one of my students at Washington University in St. Louis, Perry peng, how his team got the top position in my Kaggle In-Class competition. Their technique used a highly optimized convolution neural network that handled overfitting well and scored the best RMSE of a competition of 80+ students. Thank you, Perry and […]
1634904025_maxresdefault.jpg
154
05:26

Drop Out for Keras to Decrease Overfitting (5.4)

Dropout is a regularization that is very popular for deeplearning and keras. This technique removes a certain percentage of the neurons during each training step. Code for This Video: https://github.com/jeffheaton/t81_558_deep_learning/blob/master/t81_558_class_05_4_dropout.ipynb Course Homepage: https://sites.wustl.edu/jeffheaton/t81-558/ Follow Me/Subscribe: https://www.youtube.com/user/HeatonResearch https://github.com/jeffheaton Tweets by jeffheaton Support Me on Patreon: https://www.patreon.com/jeffheaton Source of this machine learning/AI Video
1634778370_maxresdefault.jpg
133
06:19

Using L1 and L2 Regularization with Keras to Decrease Overfitting (5.3)

L1 and L2 are classic regularization techniques that can be used in deeplearning and keras. Both techniques work by simplifying the weight connections in the neural network. Code for This Video: https://github.com/jeffheaton/t81_558_deep_learning/blob/master/t81_558_class_05_3_keras_l1_l2.ipynb Course Homepage: https://sites.wustl.edu/jeffheaton/t81-558/ Follow Me/Subscribe: https://www.youtube.com/user/HeatonResearch https://github.com/jeffheaton Tweets by jeffheaton Support Me on Patreon: https://www.patreon.com/jeffheaton Source of this machine learning/AI Video
1634774716_maxresdefault.jpg
60
08:29

9.2: Using L1 and L2 Regularization in Keras and TensorFlow (Module 9, Part 2)

Making use of L1 (ridge) and L2 (lasso) regression in Keras. Regularization helps to reduce overfitting by reducing the complexity of the weights. This video is part of a course that is taught in a hybrid format at Washington University in St. Louis; however, all the information is online and you can easily follow along. […]
1634605582_maxresdefault.jpg
316
14:17

Early Stopping in Keras to Prevent Overfitting (3.4)

It can be difficult to know how many epochs to train a neural network for. Early stopping stops the neural network from training before it begins to seriously overfitting. Generally too many epochs will result in an overfit neural network and too few will be underfit. Code for This Video: https://github.com/jeffheaton/t81_558_deep_learning/blob/master/t81_558_class_03_4_early_stop.ipynb Course Homepage: https://sites.wustl.edu/jeffheaton/t81-558/ Follow […]
1634566473_maxresdefault.jpg
1.35K
01:14:17

Intro to Deep Learning (ML Tech Talks)

An overview of Deep Learning, including representation learning, families of neural networks and their applications, a first look inside a deep neural network, and many code examples and concepts from TensorFlow. This talk is part of a ML speaker series we recorded at home. You can find all the links from this video below. I […]