Bias-Variance Trade-off - Explained

In this video we discuss why underfitting and overfitting happen in machine learning models. We look at the fundamental theory behind bias and variance, and try to derive the results from there.

Notes
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
Between 03:30 - 03:45, on the bottom card, the parenthesis should say "Low Variance", not "High Variance" as it is in the video. Sorry for any confusion that this might have caused.

Related Videos
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
Why neural networks are universal functions approximators:    • Why Neural Networks Can Learn Any Fun...  
Bagging vs Boosting:    • Bagging vs Boosting - Ensemble Learni...  
Why neural networks are deep:    • Deep by Design: Why Depth Matters in ...  
Why residual connections work:    • Why Residual Connections (ResNet) Work  
Why we need activations in neural nets:    • Why We Need Activation Functions In N...  

Contents
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
00:00 - Intro
00:28 - Bias and Variance
01:45 - Bias and Variance Example
02:35 - The Bias Variance Trade-off
03:14 - Bias and Variance in Neural Nets/Linear Regression
04:32 - Outro

Follow Me
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
🐦 Twitter: @datamlistic   / datamlistic  
📸 Instagram: @datamlistic   / datamlistic  
📱 TikTok: @datamlistic   / datamlistic  

Channel Support
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
The best way to support the channel is to share the content. ;)

If you'd like to also support the channel financially, donating the price of a coffee is always warmly welcomed! (completely optional and voluntary)
► Patreon:   / datamlistic  
► Bitcoin (BTC): 3C6Pkzyb5CjAUYrJxmpCaaNPVRgRVxxyTq
► Ethereum (ETH): 0x9Ac4eB94386C3e02b96599C05B7a8C71773c9281
► Cardano (ADA): addr1v95rfxlslfzkvd8sr3exkh7st4qmgj4ywf5zcaxgqgdyunsj5juw5
► Tether (USDT): 0xeC261d9b2EE4B6997a6a424067af165BAA4afE1a

#bias #variance #underfitting #overfitting #biasvariancetradeoff