Introduction: DeepLearning.TV is all about Deep Learning, the field of study that teaches machines to perceive the world. Starting with a series that simplifies Deep Learning, the channel fe…
- I am Not Madame Bovary / 我不是潘金莲 (2016)
- Mr. Donkey / 驴得水 (2016)
- Mr. Six / 老炮儿 (2015)
- Mountains May Depart / 山河故人 (2015)
- Gone with the Bullets / 一步之遥 (2014)
- A Touch of Sin / 天注定 (2013)
- Back to 1942 / 一九四二 (2012)
- White Deer Plain / 白鹿原 (2012)
- Let the Bullets Fly / 让子弹飞 (2010)
- Still Life / 三峡好人 (2006)
- Cell Phone / 手机 (2003)
- Devils on the Doorstep / 鬼子来了(2000)
- To Live / 活着 (1994)
- Farewell My Concubine / 霸王别姬 (1993)
I still need more time in reading and understanding MCMC and RBM. However, I’d like to share some learning materials for everyone. Hope you could find these helpful.
- An Introduction to MCMC for Machine Learning
- Reducing the Dimensionality of Data with Neural Networks
- A fast learning algorithm for deep belief nets
- A Practical Guide to Training Restricted Boltzmann Machines
- An Introduction to Restricted Boltzmann Machines
- Training restricted Boltzmann machines: An introduction
- Sparse deep belief net model for visual area V2
- Classification using Discriminative Restricted Boltzmann Machines
- Learning Invariant Representations with Local Transformations
- Unsupervised feature learning and deep learning a review and new perspectives
- Relevant literature
- To be continued …
“These violent delights have violent ends
And in their triumph die, like fire and powder,
Which, as they kiss, consume.”
Last week, I finished watching the amazing TV series Westworld. As a Sci-Fi fan, I kindly tired of the repeated laser gun and space ships, this TV series is really an open-minded setup with a solid storyline of Artificial Intelligence (Hosts) and Human (Makers/Vistors).
Imagine that technology was so advanced, you could step into a wholly immersive virtual reality world that was very nearly as real as our own. A theme park, where you can live out your fantasies. No awkward headsets or VR to wear. No fuzzy graphics, keyboard, and mice. Because it’s all real—for the most part.
You can buy drinks at the local saloon, or pay for sex with one of the AI prostitutes at the bar. You can get in gun fights with outlaws, or save the town from murderous highwaymen. Or, if you really want to, you can go to a farm house, kill the elderly parents there and rape their daughter. This is where the stories are written, tinkered with, and the microcosm of Westworld manipulated for the entertainment of its guests.
Dear friends, I’m sharing this great news that Google opens Embedding Projector for the world. Now everyone could enjoy the convenient approach to visualize high dimensional data on the web browser.
Based on my personal programming experience, t-SNE is a great method for high-dimensional data visualization, better than PCA on some data sets like MNIST.
Moreover, check the following papers if you want to learn more details about dimensionality reduction.
- L.J.P. van der Maaten and G.E. Hinton. Visualizing High-Dimensional Data Using t-SNE. Journal of Machine Learning Research 9(Nov):2579-2605, 2008. PDF [Supplemental material] [Talk]
- L.J.P. van der Maaten, E.O. Postma, and H.J. van den Herik. Dimensionality Reduction: A Comparative Review. Tilburg University Technical Report, TiCC-TR 2009-005, 2009. PDF
Here are the links for more information:
- The Original Blog of Google Research: https://research.googleblog.com/2016/12/open-sourcing-embedding-projector-tool.htm
- Embedding Projector: http://projector.tensorflow.org/ (Note: when running MNIST, it may slow down your computer, be careful to this).
- Related Technique of t-SNE: http://distill.pub/2016/misread-tsne/
I am not sure when I could finish these, but I will try.
Just sharing my sources of Deep learning, if anyone finds this post helpful, please share it. The content will be updated with the new techniques and information.
AI, Machine Learning and Deep Learning are transforming numerous industries. But building a machine learning system requires that you make practical decisions:
- Should you collect more training data?
Should you use end-to-end deep learning?
How do you deal with your training set not matching your test set?
and many more.
Historically, the only way to learn how to make these “strategy” decisions has been a multi-year apprenticeship in a graduate program or company. I am writing a book to help you quickly gain this skill so that you can become better at building AI systems.
The book will be around 100 pages and contain many easy-to-read 1-2 page chapters. If you would like to receive a draft of each chapter as it is finished, please sign up for the mailing list.