t-distributed Stochastic Neighbor Embedding (t-SNE) | Dimensionality Reduction Techniques (4/5)

To try everything Brilliant has to offer-free-for a full 30 days, visit brilliant.org/DeepFindr. The first 200 of you will get 20% off Brilliant’s annual premium subscription.
(Video sponsered by Brilliant.org)
▬▬ Papers / Resources ▬▬▬
Colab Notebook: colab.research.google.com/dri...
Entropy: gregorygundersen.com/blog/202...
Attractive / Repulsive Forces Gradient: jmlr.org/papers/volume23/21-0...
t-SNE Parameters distill: distill.pub/2016/misread-tsne/
Other great resources:
- By the t-SNE author: lvdmaaten.github.io/tsne/
- A good view on probability: siegel.work/blog/tSNE/
- CalTech tutorial: bebi103.caltech.edu.s3-website...
- Great visuals: newsletter.theaiedge.io/p/for...
- SNE vs T-SNE: / visualization-method-s...
- t-SNE in raw numpy: nlml.github.io/in-raw-numpy/i...
- t-SNE in raw javascript: observablehq.com/@nstrayer/t-...
- Video by the t-SNE author: • CVPR18: Tutorial: Part...
Image Sources:
- Perplexity image: stats.stackexchange.com/quest...
▬▬ Support me if you like 🌟
►Link to this channel: bit.ly/3zEqL1W
►Support me on Patreon: bit.ly/2Wed242
►Buy me a coffee on Ko-Fi: bit.ly/3kJYEdl
►E-Mail: [e-mail]
▬▬ Used Music ▬▬▬▬▬▬▬▬▬▬▬
Music from #Uppbeat (free for Creators!):
uppbeat.io/t/sulyya/weather-c...
License code: ZRGIWRHMLMZMAHQI
▬▬ Used Icons ▬▬▬▬▬▬▬▬▬▬
All Icons are from flaticon: www.flaticon.com/authors/freepik
▬▬ Timestamps ▬▬▬▬▬▬▬▬▬▬▬
00:00 Intro
00:30 Manifold learning
02:40 Relevant Papers & Agenda
03:25 Stochastic Neighbor Embedding (SNE)
03:56 Pairwise distances
04:35 Distance to Probability
06:06 Conditional Probability Math
07:05 Adjustment of Variance
08:20 Perplexity
09:55 How to find the variance
11:15 KL-divergence
12:55 Shepard Diagram
13:15 Gradient and it's interpretation
14:15 N-body simulation
14:35 Full SNE Algorithm
15:15 t-distributed Stochastic Neighbor Embedding (t-SNE)
15:28 Crowding Problem and how to solve it
17:58 Gaussian vs. Student's t Distribution
19:21 Symmetric Probabilities
20:35 Early Exaggeration
22:50 SNE vs. t-SNE
23:08 Brilliant.org Sponsoring
24:14 Code
27:15 Distill.pub Blogpost
27:49 Barnes-Hut t-SNE
29:54 Comparison
31:06 Outro
▬▬ My equipment 💻
- Microphone: amzn.to/3DVqB8H
- Microphone mount: amzn.to/3BWUcOJ
- Monitors: amzn.to/3G2Jjgr
- Monitor mount: amzn.to/3AWGIAY
- Height-adjustable table: amzn.to/3aUysXC
- Ergonomic chair: amzn.to/3phQg7r
- PC case: amzn.to/3jdlI2Y
- GPU: amzn.to/3AWyzwy
- Keyboard: amzn.to/2XskWHP
- Bluelight filter glasses: amzn.to/3pj0fK2

[@DeepFindr]

To try everything Brilliant has to offer-free-for a full 30 days, visit brilliant.org/DeepFindr​. The first 200 of you will get 20% off Brilliant’s annual premium subscription.

[@chemicalengineeringfriends217]

Great videos! Looking forward to other parts :)

[@clairenajjuuko7664]

Great video. looking forward to the UMAP video. Will you also be doing something on FAMD?

[@DeepFindr]

Thanks! So far only UMAP is planned but maybe more methods will be added in the future :)

[@lucapalese475]

Really nice! I will read those papers , I guess the backprop is more complex with the t-distribution

[@DeepFindr]

Actually it should be easier because the distribution has an easier function