CS 412 • Final • Machine Learning
Eğitmen
Nursena Köprücü Aslan
PhD in Computer Science
Koç Üniversitesi’nde Bilgisayar Mühendisliği okudum ve aynı zamanda Matematik alanında çift anadal yaptım. Ardından Imperial College London’da Machine Learning and Artificial Intelligence alanında yüksek lisansımı tamamladım. Şu anda University of Cambridge'te doktora çalışmalarımı sürdürüyorum.
Paketi Tamamla
🎓 Sabancı Üniversitesi öğrencilerinin %92'si tüm paketi alarak çalışıyor.
Konular
Probability Review
Counting and Probability
Conditional Probability and Independence
Bayes' Rule
Discrete Random Variables
Continuous Random Variables
Expected Value and Variance
Bernoulli and Binomial Distributions
Continuous Uniform Distribution
Exponential Distribution
Normal Distribution
Laplace and Logistic Distributions
Naive Bayes
Naive Bayes Approach
Curse of Dimensionality
Bayes Classifier vs. Naive Bayes
Independence & Conditional Independence
Naive Bayes Classification
How Naive Bayes Simplifies Parameter Estimation
Example with categorical variables
Naive Bayes Subtleties
Bayes Classifier with Multivariate Normal Distribution
Modeling Multivariate Data: Estimation, Normal Distributions, and Naive Bayes
Discrete Features & Multivariate Regression
Sample Mean and Covariance Matrix
Mahalanobis Distance
Naive Bayes Classification (Discrete Features)
Parametric Classification and Estimation
Maximum Likelihood Estimation(MLE)
Bernoulli Likelihood
Multinomial Likelihood and Smoothing
Parametric Classification
Unequal Variances → Quadratic Boundary
Gaussian Classification Boundary
Multivariate Classification: Linear, Quadratic, and Model Selection
LDA vs QDA Classification
Text Classification w/ Naive Bayes
Text Classification with Naïve Bayes
Text Classification Tasks
Bag-of-Words (BoW)
Naive Bayes scoring with BoW
Bernoulli Naive Bayes
Multinomial Naive Bayes
Multinomial Naive Bayes - 2
Laplace smoothing
Practical Notes
Text Processing
Why raw term counts aren’t enough
Transforming TF
IDF
TF-IDF
Text Preprocessing
Tokenization and Token normalization
Stop words + Stemming vs Lemmatization
Word Embeddings
Why text is harder than “normal” ML inputs
Bag of Words / TF-IDF
One-Hot Encoding & Distributional Hypothesis
Word Embeddings
Word2Vec idea & CBOW vs Skip-gram
Training objective & Negative Sampling
Scoring function
What embeddings capture
Limitations of Word2Vec embeddings
Language Models
Word2Vec Training Data
Semi/Self-Supervised Learning
Semi/Self-Supervised Learning
Pretext & Pseudo-labeling
Consistency Regularization
Representation Learning & Contrastive Learning
Semi-Supervised Learning & Entropy Minimization
Early Work in SSL
Clustering
Introduction and Mixture Densities
K-Means Clustering
Sample Final Questions I
Pass Rates & Majors (Bayes; Law of Total Probability)
Weighted Least Squares (Closed-Form Solution, Matrix View & Interpretation)
MLE for α (positive support, exponential tail)
Linear Discriminant with Equal Variance
MLP with Hard-Threshold Units
Should we initialize all MLP weights to zero?
One Shared Network vs. Three Separate Networks
Naive Histogram Estimator vs. Parzen Windows (Kernel)
Kernel Smoother
Naive Density Estimator (Bandwidth effect & validity)
Comparing Two Splits (Gini vs. Misclassification)
Prepruning vs. Postpruning (Which and Why?)
Sample Final Questions II
Why Not Regression for Classification?
From Binary to Multiclass: One-vs-All / One-vs-One with a Binary Classifier
Max-shift for SoftMax
Why Initialize Weights Near Zero?
Adaptive Learning Rates in Gradient Descent
When Do Direct Input Output Links Help in an MLP?
Mahalanobis vs. Euclidean: Why and When?
Discrete Attribute in Decision Trees
Regularized Least Squares
Gaussian Generative Model → Logistic Posterior
Naive Bayes Text Classification with Binary Features
Derivative of Softmax
Choosing Between Two Splits: Gini vs. Misclassification
Sample Final Questions III
Single-Neuron Sigmoid + MSE
Decision Trees: Gini Impurity Split Comparison
Decision Trees: Entropy & Information Gain Split Comparison
MLE for a Discrete PMF
1-NN LOOCV on Patient Dataset
Linear Regression + MSE: Gradient Descent Step Size Effects
Past Exam Questions
Mean Square Error for Linear Regression
Gradient Descent Update
k-NN Regression Prediction
Decision Boundary and Building a Network for Binary Classification
Derivative of Squared Error
Computing Input and Output of a Convolution Node
True/False Reasoning on Activation, Linear Networks, and Gradient Descent
Computing Total Probability
True/False on Scaling, k-NN, Intrinsic Error and Model Complexity
Output Size of a Conv Layer
Regression: Test-Set MSE
Generalization & Overfitting: True/False
Baseline Error: ZeroR vs Random Guessing
Entropy: Fair Die & Bias Effect
Decision Trees: ID3 Optimality + Key Advantage
Decision Tree Split: Remaining Entropy
Değerlendirmeler
Sıkça Sorulan Sorular
Örneğin, Koç Üniversitesi - MATH 101 (Calculus) veya başka bir okulun benzer dersi olsun, paketlerimiz tam da o derse göre tasarlanır. Böylece nokta atışı çalışır, zaman kazanırsın.
Sınava özel videolar —konu anlatımları, çıkmış sorular ve çözümleri, özet notlar—içerir. Sınavda sıkça çıkan soruları hedefler. Eğitmenlerimiz, üniversitenin akademik takvimini takip ederek paketleri sürekli günceller. Böylece, gereksiz detaylarla vakit kaybetmeden başarını artırmaya odaklanabilirsin.
Ders İçeriği
Probability Review
Counting and Probability
Conditional Probability and Independence
Bayes' Rule
Discrete Random Variables
Continuous Random Variables
Expected Value and Variance
Bernoulli and Binomial Distributions
Continuous Uniform Distribution
Exponential Distribution
Normal Distribution
Laplace and Logistic Distributions
Naive Bayes
Naive Bayes Approach
Curse of Dimensionality
Bayes Classifier vs. Naive Bayes
Independence & Conditional Independence
Naive Bayes Classification
How Naive Bayes Simplifies Parameter Estimation
Example with categorical variables
Naive Bayes Subtleties
Bayes Classifier with Multivariate Normal Distribution
Modeling Multivariate Data: Estimation, Normal Distributions, and Naive Bayes
Discrete Features & Multivariate Regression
Sample Mean and Covariance Matrix
Mahalanobis Distance
Naive Bayes Classification (Discrete Features)
Parametric Classification and Estimation
Maximum Likelihood Estimation(MLE)
Bernoulli Likelihood
Multinomial Likelihood and Smoothing
Parametric Classification
Unequal Variances → Quadratic Boundary
Gaussian Classification Boundary
Multivariate Classification: Linear, Quadratic, and Model Selection
LDA vs QDA Classification
Text Classification w/ Naive Bayes
Text Classification with Naïve Bayes
Text Classification Tasks
Bag-of-Words (BoW)
Naive Bayes scoring with BoW
Bernoulli Naive Bayes
Multinomial Naive Bayes
Multinomial Naive Bayes - 2
Laplace smoothing
Practical Notes
Text Processing
Why raw term counts aren’t enough
Transforming TF
IDF
TF-IDF
Text Preprocessing
Tokenization and Token normalization
Stop words + Stemming vs Lemmatization
Word Embeddings
Why text is harder than “normal” ML inputs
Bag of Words / TF-IDF
One-Hot Encoding & Distributional Hypothesis
Word Embeddings
Word2Vec idea & CBOW vs Skip-gram
Training objective & Negative Sampling
Scoring function
What embeddings capture
Limitations of Word2Vec embeddings
Language Models
Word2Vec Training Data
Semi/Self-Supervised Learning
Semi/Self-Supervised Learning
Pretext & Pseudo-labeling
Consistency Regularization
Representation Learning & Contrastive Learning
Semi-Supervised Learning & Entropy Minimization
Early Work in SSL
Clustering
Introduction and Mixture Densities
K-Means Clustering
Sample Final Questions I
Pass Rates & Majors (Bayes; Law of Total Probability)
Weighted Least Squares (Closed-Form Solution, Matrix View & Interpretation)
MLE for α (positive support, exponential tail)
Linear Discriminant with Equal Variance
MLP with Hard-Threshold Units
Should we initialize all MLP weights to zero?
One Shared Network vs. Three Separate Networks
Naive Histogram Estimator vs. Parzen Windows (Kernel)
Kernel Smoother
Naive Density Estimator (Bandwidth effect & validity)
Comparing Two Splits (Gini vs. Misclassification)
Prepruning vs. Postpruning (Which and Why?)
Sample Final Questions II
Why Not Regression for Classification?
From Binary to Multiclass: One-vs-All / One-vs-One with a Binary Classifier
Max-shift for SoftMax
Why Initialize Weights Near Zero?
Adaptive Learning Rates in Gradient Descent
When Do Direct Input Output Links Help in an MLP?
Mahalanobis vs. Euclidean: Why and When?
Discrete Attribute in Decision Trees
Regularized Least Squares
Gaussian Generative Model → Logistic Posterior
Naive Bayes Text Classification with Binary Features
Derivative of Softmax
Choosing Between Two Splits: Gini vs. Misclassification
Sample Final Questions III
Single-Neuron Sigmoid + MSE
Decision Trees: Gini Impurity Split Comparison
Decision Trees: Entropy & Information Gain Split Comparison
MLE for a Discrete PMF
1-NN LOOCV on Patient Dataset
Linear Regression + MSE: Gradient Descent Step Size Effects
Past Exam Questions
Mean Square Error for Linear Regression
Gradient Descent Update
k-NN Regression Prediction
Decision Boundary and Building a Network for Binary Classification
Derivative of Squared Error
Computing Input and Output of a Convolution Node
True/False Reasoning on Activation, Linear Networks, and Gradient Descent
Computing Total Probability
True/False on Scaling, k-NN, Intrinsic Error and Model Complexity
Output Size of a Conv Layer
Regression: Test-Set MSE
Generalization & Overfitting: True/False
Baseline Error: ZeroR vs Random Guessing
Entropy: Fair Die & Bias Effect
Decision Trees: ID3 Optimality + Key Advantage
Decision Tree Split: Remaining Entropy