CSCI B 609 Foundations of Data Science Lecture
CSCI B 609: “Foundations of Data Science” Lecture 10/11: Random Walks and Markov Chains + ML Intro Slides at http: //grigory. us/data-science-class. html Grigory Yaroslavtsev http: //grigory. us
Project Example: Gradient Descent in Tensor. Flow • Gradient Descent (will be covered in class) • Adagrad: http: //www. magicbroom. info/Papers/Duchi. Ha. Si 10. pdf • Momentum (stochastic gradient descent + tweaks): http: //www. cs. toronto. edu/~hinton/absps/naturebp. pdf • Adam (Adaptive + momentum): http: //arxiv. org/pdf/1412. 6980. pdf • FTRL: http: //jmlr. org/proceedings/papers/v 15/mcmahan 11 b/mc mahan 11 b. pdf • RMSProp: http: //www. cs. toronto. edu/~tijmen/csc 321/slides/lecture_ slides_lec 6. pdf
Random Walks and Markov Chains •
Strongly Connected Components •
Matrix Form and Stationary Distribution •
Stationary Distribution •
Stationary Distribution Theorem •
Stationary Distribution Theorem Cont. •
Fundamental Theorem of Markov Chains •
Fundamental Theorem of Markov Chains •
Intro to ML •
Intro to ML •
Overfitting and Uniform Convergence •
Examples •
Online Learning + Perceptron Algorithm •
Perceptron Algorithm •
Perceptron Analysis cont. •
Perceptron with noisy data •
Proof of noisy perceptron •
- Slides: 19