Differences

This shows you the differences between two versions of the page.

--- course_outline [2012/08/30 18:50] – hj
+++ course_outline [2016/09/12 21:11] (current) – hj
@@ Line 3: / Line 3: @@
 The course outline is a guideline to topics that will be discussed in the course, and when they will be discussed:
-===== Week 1 =====
+===== Weeks 1 and 2 =====
-**Introduction**: application background; a big picture; speech sounds; spoken language
+**Machine Learning** (basic concepts)
-(reading assignment)
-===== Week 2 =====
+**Math Background**: probabilities; Bayes theorem; statistics; estimation; regression; hypothesis testing; entropy; mutual information; decision tree; optimization theory and convex optimization; matrix decomposition
-**Math Background**: probabilities; Bayes theorem; statistics; estimation; regression; hypothesis testing; entropy; mutual information; decision tree; optimization theory and convex optimization
 ===== Week 3 =====
-**Pattern Classification**: pattern classification & pattern verification; Bayesian decision theory
+**Data, Feature and Model**:  Feature Engineering, Feature Extraction (PCA, LDA, etc), Data Virtualization
 ===== Week 4 =====
-**Generative Models**: model estimation: maximum likelihood, Bayesian learning, EM algorithm; multivariate Gaussian, Gaussian mixture model, Multinomial, Markov Chain model, etc.
+**Machine Learning (ML) Basics**:  Learnability; Some basic ML Concepts; Bayesian decision theory
 ===== Week 5 =====
-**Discriminative Models**: Linear discriminant functions; support vector machine (SVM); large margin classifiers; sparse kernel machines; Neural networks
+**Generative Models (I)**: model estimation: maximum likelihood, Bayesian learning, multivariate Gaussian, Multinomial, Markov Chain model, etc.
 ===== Week 6 =====
-**Hidden Markov Model (HMM)**: HMM vs. Markov chains; HMM concepts; Three algorithms: forward-backward; Viterbi decoding; Baum-Welch learning.
+**Discriminative Models (I) **: Linear discriminant functions; support vector machine (SVM); large margin classifiers; sparse kernel machines;
-===== Week 7  =====
-midterm presentation
-===== Week 8 =====
+===== Week 7 =====
-**Automatic Speech Recognition (ASR) (I)**: Acoustic and Language Modeling:
+**Discriminative Models (II) **:  Neural Networks and Deep Learning
-HMM for ASR;  ASR as an example of pattern classification;
-Acoustic modeling: HMM learning (ML, MAP, DT); parameter tying (decision tree based state tying); n-gram models: smoothing, learning, perplexity, class-based.
-===== Week 9 =====
+===== Week 8  =====
-**Automatic Speech Recognition (ASR) (II)**: Search - why search; Viterbi decoding in a large HMM; beam search; tree-based lexicon; dynamic decoding; static decoding; weighted finite state transducer (WFST)
+**Generative Models (II)**: EM algorithm; Finite Mixture models, e.g. Gaussian mixture model
-===== Week 10 =====
+===== Week 9  =====
-**Spoken Language Processing (I)**: text categorization
+**Generative Models (III)**:  graphical models, directed vs. indirected graphical models, exact inference, approximate inference (loopy belief propagation, variational inference, Monte Carlo methods)
-classify text documents: call/email routing, topic detection, etc.
-vector-based approach, Naïve Bayes classifier; Bayesian networks, etc.
-(2) HMM applications: Statistical Part-of-Speech (POS) tagging;
- Language understanding: hidden concept models.
-===== Week 11 =====
+===== Week 10  =====
-**Spoken Language Processing (II)**: statistical machine translation
+**Selected Advanced Topics**: Hidden Markov Model (HMM), HMM concepts; Three algorithms: forward-backward; Viterbi decoding; Baum-Welch learning.
- IBM’s models for machine translation: lexicon model, alignment model, language model
- training process, generation & search
+===== Weeks 11-12 =====
-===== Week 12 =====
 student presentation