Lectures: ML and DLC

If it is your first time working with Machine Learning then we highly recommend you start with Part 1 ✨.

However, if you are comfortable with the basics, just skip ahead to the introduction of computer vision and DeepLabCut

Part 1: Introduction to Machine Learning (50 min)

Note

Art Samuel stated that Machine Learning “is the field of study that gives computers the ability to learn without being explicitly programmed.”

Introduction to Machine Learning by Eric Grimson

This lecture is part of the MIT class —- Introduction to Computational Thinking and Data Science, Fall 2016

Content of this lecture:

• What is machine learning? (minute 7+)

• Supervised & Unsupervised learning (minute 14+)

• Training & test performance (minute 48+)

Fig. 1 Traditional Programming vs. Machine Learning from Grimson’s lecture.

If you want to expand your knowledge, check out these materials:

• Chapter 1 of Deep Learning by Ian Goodfellow, Yoshua Bengio and Aaron Courville

• The basics of neural networks, and the math behind how they learn by 3Blue1Brown

• Neural Networks and Deep Learning by Michael Nielsen

• Understanding Mini-Batch Gradient Descent by Andrew Ng (DeepLearningAI)

• For a clarification of the differences of machine learning (ML), deep learning (DL), and artificial intelligence (AI) refer to Figure 1.4 in the Deep Learning book.

Part 2: Introduction to Computer Vision and DeepLabCut

Introduction to Computer Vision and DeepLabCut by Alexander Mathis

For further information, check out the key papers from this lecture: [RDS+15], [HZRS16], [MMC+18], [MSLM20], [MBS+21].

If you want to expand your knowledge, check out these materials:

• Brief history of vision and computer vision by Fei-Fei Li

• ResNet paper

• Part II of Deep Learning by Ian Goodfellow, Yoshua Bengio and Aaron Courville

• Lecture collection to deep dive into details of the deep learning architectures from Stanford University

References

[HZRS16]

Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition, 770–778. 2016.

[MBS+21]

Alexander Mathis, Thomas Biasi, Steffen Schneider, Mert Yuksekgonul, Byron Rogers, Matthias Bethge, and Mackenzie W Mathis. Pretraining boosts out-of-domain robustness for pose estimation. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, 1859–1868. 2021.

[MMC+18]

Alexander Mathis, Pranav Mamidanna, Kevin M. Cury, Taiga Abe, Venkatesh N. Murthy, Mackenzie Weygandt Mathis, and Matthias Bethge. DeepLabCut: markerless pose estimation of user-defined body parts with deep learning. Nature Neuroscience, 21(9):1281–1289, August 2018. URL: https://doi.org/10.1038/s41593-018-0209-y, doi:10.1038/s41593-018-0209-y.

[MSLM20]

Alexander Mathis, Steffen Schneider, Jessy Lauer, and Mackenzie Weygandt Mathis. A primer on motion capture with deep learning: principles, pitfalls, and perspectives. Neuron, 108(1):44–65, October 2020. URL: https://doi.org/10.1016/j.neuron.2020.09.017, doi:10.1016/j.neuron.2020.09.017.

[RDS+15]

Olga Russakovsky, Jia Deng, Hao Su, Jonathan Krause, Sanjeev Satheesh, Sean Ma, Zhiheng Huang, Andrej Karpathy, Aditya Khosla, Michael Bernstein, and others. Imagenet large scale visual recognition challenge. International journal of computer vision, 115(3):211–252, 2015.