Course Syllabus

Below is a syllabus template that includes WSU's required syllabus elements. Please complete all items highlighted in yellow. 

 

 Statistical Foundations in Artificial Intelligence

DATA/STAT 460

Semester and Year [tbd]

Number of Credit Hours 3

Prerequisites Data 219, Data 324; Math 216; Stat/Data 360

Course Details

Day and Time: [tbd]

Meeting Location: [tbd]

 

Instructor Contact Information

Instructor Name: [tbd]

Instructor Contact Information: [office location, phone, email] [tbd]

Instructor Office Hours: [click here for best practices] [tbd]

 

TA Name: [tbd]

TA Contact Information: [office location, phone, email]: [tbd]

TA Office Hours: [click here for best practices] [tbd]

 

Course Description

This is a 3-credit hour, face-to-face course, appropriate for seniors in Mathematics, Statistics, and Data Analytics. The course will use several reference books and instructor prepared materials, and it will cover the following topics as statistical foundations of AI: 

• Principles and ethics of collecting, generating, handling, and processing data.
• Probability theory for generative models for data processing, visualization, and modeling.
• Fundamentals of statistical inference, predictive modeling, and model validation, and resampling based statistical methods.
• Fundamentals of statistical learning, decision theoretic formulation of learning, regularization techniques, ensemble methods, and gradient-based learning.
• Ethics of AI, and specific deep learning models and their AI applications.

 

Course Materials 

Books: None

Reference books:    

• “Deep Learning” by Christopher M. Bishop
• “Statistical Learning Theory” by Vladimir N. Vapnik
• “Statistical Inference” by  George Casella, Roger Berger

Fees: None

Student Learning Outcomes (SLOs) 

1. Data Collection, Generation and Handling in AI:
   • Identify, represent, and manage various data types.
   • Ensure data quality and ethics in data collection and handling.
2. Probability Theory for AI:
   • Apply fundamental probability concepts and probability distributions.
   • Understand the Law of Large Numbers and Central Limit Theorem.
3. Statistical Inference and Resampling Methods:
   • Perform estimation and hypothesis testing.
   • Use Bayesian inference
   • Use resampling based methods for inference.
4. Predictive Modeling:
   • Develop and evaluate regression and classification models.
   • Implement regularization techniques and non-parametric methods.
   • Use resampling based methods for model validation and performance evaluation.
5. Statistical Learning and AI:
   • Understand decision theoretic formulation of learning.
   • Understand principles of machine learning and AI.
   • Understand gradient based learning.
   • Understand some key models of machine learning and AI.
   • Apply knowledge to real-world problems, considering ethical implications.

Assessment of Learning Outcomes

Students’ learning outcomes will be accessed by a combination of homework assignments, exams, and projects. Detailed information is given by the following.

Mid-term exam: there will be a mid-term exam.

Make-up exams:  Make-up exams will be allowed on a case-by-case basis and will be given to accommodate university conflicts, illness or other unforeseen emergencies.  Students must let the instructor know, as soon as possible, that they will not be able to take the scheduled exam. Make-up exams must be completed before the WSU official final exam date(s) for the semester of the course and within a reasonable period after they were originally scheduled.

Homework:  Approximately 5 homework assignments will be given during the semester.  These will come from problems provided by reference books or materials discussed in the lectures. Homework assignments will primarily consist of methodological and programming exercises. Please submit answers to HW assignments with necessary supporting computer codes, and organize them. Late homework will only be accepted under extenuating circumstances, such as an extended illness. 

Written Projects:  One final project will be assigned during the semester. Each project will consist of 4 components: (1) a typed write-up that contains 5 components:  introduction, methods used to conduct the analysis, results of the analysis, conclusions from and discussion on the analysis, and a reference section; (2) computer codes used to conduct the analysis; (3) most relevant outputs from the analysis, which can be incorporated in component (1); (4) a detailed description of the contribution of each member of the group towards the project.  Each project can be completed by up to 2 students.

 

---

Course Schedule

[Please note that a WSU semester is 15 weeks + Thanksgiving/Spring Break. The schedule below does not include the break.]

Dates	Lesson Topic	Objective	Topics
Week 1 & 2 [dates]	Fundamentals of Data collection, generation and management	Introduce statistical principles and ethnic principles of data handling.	Introduction to Data Types of Data: Quantitative, Qualitative, Structured, Unstructured. Data Representation: Tables, Graphs, Curves, etc. Scales of Measurement: Nominal, Ordinal, Interval, Ratio. Population and Sampling Population vs. Sample. Sampling Techniques: Random, Stratified, Cluster, Systematic. Sampling Bias and Errors. Stability and reproducibility for AI Data Collection and Management Data Extraction, Generation, and Collection. Survey Design and Questionnaires. Introduction to Databases and Data Storage Formats (SQL, NoSQL). Ethical Considerations in Data Collection. Data Cleaning and Preprocessing Handling Missing Data and Measurement Error in Data. Data Transformation, Scaling, Normalization, and Interpolation. Outlier Detection and Treatment.
Week 3 & 4 [dates]	Probability and Random Variables	Build a strong foundation in probability theory for AI applications.	Probability Basics Definitions: Chances, and Axiomatic Probability. Conditional Probability and Bayes’ Theorem. Independence of Events. Random Variables and Expectations Discrete Random Variable, Continuous Random Variables. Probability Mass Function (PMF) and Probability Density Function (PDF). Expectation, Variance, Covariance, and Moments. Common Probability Distributions Discrete and Combinatoric: Bernoulli, Binomial, Geometric, Poisson. Continuous: Uniform, Normal, Exponential, Gamma, Beta. Multivariate Distributions: Multinomial, Multivariate Normal, Multivariate Student t. Law of Large Numbers and Central Limit Theorem Implications in Statistical Learning Implications in AI and Machine Learning.
Week 5 - 7 [dates]	Statistical Inference and Resampling Methods	Provide tools to draw conclusions about populations from data.	Estimation Point Estimation: Properties of Estimators (Bias, Consistency, Efficiency). Interval Estimation: Confidence Intervals, Credible Intervals. Hypothesis Testing Null and Alternative Hypotheses. Type I and Type II Errors. p-values, q-values, e-values, and Significance Levels. Parametric and Non-Parametric Tests. Bayesian Inference Prior, Conditional, Posterior, and Evidence. Markov Chain Monte Carlo (MCMC) Methods. Application to Bayesian Learning and Bayesian Neural Networks Resampling techniques Bootstrap and Cross-Validation. Permutation Techniques Applications in Confidence Intervals and Hypothesis Testing. Applications in evaluating AI Models
Week 8 - 10 [dates]	Regression and Predictive Modeling	Teach regression analysis and predictive modeling techniques.	Linear Regression Simple and Multiple Linear Regression. Assumptions and Diagnostics. Generalized Linear Models (GLMs) Logistic Regression for Binary Classification. Poisson Regression for Count Data. Diagnostics for GLMs Non-Parametric Regression Kernel Regression. Smoothing Splines and LOESS. Regularization and Shrinkage Techniques Ridge Regression and Lasso. Empirical Bayes methods. Classification Techniques k-Nearest Neighbors (kNN). Gaussian Mixture Models. Support Vector Machines (SVMs).
Week 11-13 [dates]	Statistical Learning and Special Topics	Dive deeper into the intersection of statistics and AI.	Introduction to Machine Learning Supervised vs. Unsupervised Learning. Bias-Variance Tradeoff. Decision Theoretic Formulation of Learning Learning via Empirical Risk Minimization Conformal Prediction and Conformal Learning.       Learnability Weak Learners and Strong Learners Complexity and VC-dimension NP Hardness vs Non-No Hardness Ensemble Methods Boosting Bagging and Random Forests Gradient-based Learning Algorithms Classic Gradient-based Learning. Stochastic Gradient and Implicit Regularization. Dimension Reduction and Manifold Learning Independent Component Analysis. Principal Component Analysis. Manifold Learning. Graphs and Networks Networks as Graphs. Random Graphs and Graph Laplacian. Simple Random Processes on Random Graphs.
Week 14-16 [dates]	Case Studies and Applications	Learn popular deep learning methods and gain insights through real-world examples.	Specific Deep Learning Methods Convolutional Neural Networks Generative Adversarial Networks Graphical Neural Networks. Transformers. Audoencoders. AI Applications in Data Science Classification of objects (with applications in computer vision, security system, and disease diagnosis, etc). Predictive learning (with applications in financial risk management, environmental science, language processing, agriculture, etc). Ethics and Bias in AI Statistical Fairness in AI Models. Privacy protection in AI models. Transparency in AI Models. Capstone Project Apply the learned concepts to solve a real-world problem.

 

 

Expectations for Student Effort 

You are expected to spend a minimum of 9 hours per week for a three-credit course, of which 3 hours are spent on instructor-led activities (lectures and discussions) and 6 hours are spent on outside activities. These outside activities include, but are not limited to: reading, studying, problem solving, writing, homework, and other preparations for the course. Achievement of course goals may require more than the minimum time commitment. For the most accurate and up to date information go to Academic Regulations.

 

Grading [add more lines if necessary]

Assignment Breakdown

Type of Assignment (tests, papers, etc)	Percent of Overall Grade
Homework	50%
Mid-term exam	20%
Final project	30%
Total	100%

 

Grading Schema

Grade	Percent	Grade	Percent
A	93% - 100%	C	73% - 76.99%
A-	90% - 92.99%	C-	70% - 72.99%
B+	87% - 89.99%	D+	66% - 69.99%
B	83% - 86.99%	D	60% - 65.99%
B-	80% - 82.99%	F	0% - 59.99%
C+	77% - 79.99%

 

Written Project Grading Rubric:

Introduction:  This section must consist of a description of the problem to be addressed, a concise review of pertinent literature related to the problem, and a statement of the proposed solution to the problem.  It should usually not exceed 2 pages in length.

Methods: This section should describe how the problem described in Introduction is solved. It usually is the main part of the project report.

Results: This section should present the main results of the solution to the problem and their interpretations.

Discussion: This section should discuss aspects of the solution that can be further improved, relevant issues not touched by the solution, and other problems to which the solution might be applied.

---

Attendance and Make-Up Policy 

In order to properly gain a working knowledge of the materials, attendance and participation in class are necessary.  While this will not be graded, enrolled students are highly encouraged to attend the class.

 

---

Academic Integrity Statement

You are responsible for reading WSU's Academic Integrity Policy, which is based on Washington State law. If you cheat in your work in this class you will:

-Be reported to the Center for Community Standards

-Have the right to appeal my decision

-Not be able to drop the course of withdraw from the course until the appeals process is finished

If you have any questions about what you can and cannot do in this course, ask me.

If you want to ask for a change in my decision about academic integrity, use the form at the Center for Community Standards website. You must submit this request within 21 calendar days of the decision.