Subject Datasheet
Download PDFI. Subject Specification
1. Basic Data
1.1 Title
Artificial intelligence
1.2 Code
BMEEOFTDT85
1.3 Type
Module with associated contact hours
1.4 Contact hours
| Type | Hours/week / (days) |
| Lecture | 2 |
1.5 Evaluation
Exam
1.6 Credits
3
1.7 Coordinator
| name | Béla Paláncz |
| academic rank | Professor emeritus |
| palancz.bela@emk.bme.hu |
1.8 Department
Department of Geodesy and Surveying
1.9 Website
1.10 Language of instruction
english
1.11 Curriculum requirements
Ph.D.
1.12 Prerequisites
Required previous subjects (need to be completed to register): Corresponding MSc subjects
1.13 Effective date
1 September 2022
2. Objectives and learning outcomes
2.1 Objectives
Practical and theoretical introduction to models and methods of machine and deep learning.
2.2 Learning outcomes
Upon successful completion of this subject, the student:
A. Knowledge
- General, solid knowledge of the ML and DL techniques and their applications.
B. Skills
- Finding the proper methods for the actual problem
- Recognition the advantages and handicaps of the applied methods
- Providing alternative solutions
- Ability to evaluate real project
- Select and using appropriate software
C. Attitudes
D. Autonomy and Responsibility
2.3 Methods
Lectures, electronic hand-outs, computer solution of practical problems
2.4 Course outline
| Hét | Előadások és gyakorlatok témaköre |
| 1. | Introduction. Artificial intelligence - Machine learning - Deep Learning. Tasks - Models - Learning Methods |
| 2. | Methods of feature reduction 1: PCA -SVD - KL Decomposition - TLS - IC Analysis |
| 3. | Methods of feature reduction 2: DFT - DWT - RBF Approximation - Autoencodig - Fractal Compression |
| 4. | Classification: KNN - Logistic regression - Tree Based Models - SVM - Naive Bayes Classifier |
| 5. | Clustering: KMeans - Hierarchical - Density Based Spacial - Sprectal |
| 6. | Regression: KNN - Linear - Non-Linear - Robust - Symbolic - SVM |
| 7. | Neural Networks Basic: Single and Multilayer Perceptron |
| 8. | Hopfield Net and its applications |
| 9. | Unsupervised Net and its applications |
| 10. | Recurrent Network |
| 11. | Features of Deep Learning - applications |
| 12. | Convolutional Neural Network with applications |
| 13. | Project work - consultation |
| 14. | Project work - consultation |
The above programme is tentative and subject to changes due to calendar variations and other reasons specific to the actual semester. Consult the effective detailed course schedule of the course on the subject website.
2.5 Study materials
Textbooks:
- Awange - Paláncz - Lewis - Völgyesi: Mathematical Geosciences 2nd edition, Springer 2022
- Awange - Paláncz - Völgyesi: Hybrid Imaging and Visualization Springer 2020
Online materials:
- Electronic Lecture Notes
2.6 Other information
Website:
- www.wolframcloud.com/obj/palancz/Published/Artificial_Intelligence.nb
2.7 Consultation
The instructors are available for consultation during their office hours, as advertised on the department website. Special appointments can be requested via e-mail: oktato@mail.bme.hu
This Subject Datasheet is valid for:
Inactive courses
II. Subject requirements
Assessment and evaluation of the learning outcomes
3.1 General rules
3.2 Assessment methods
| Teljesítményértékelés neve (típus) | Jele | Értékelt tanulási eredmények |
| A.1; B.1-B.5; C.1; D.1 |
The dates of deadlines of assignments/homework can be found in the detailed course schedule on the subject’s website.
3.3 Evaluation system
| Jele | Részarány |
| Összesen | 100% |
3.4 Requirements and validity of signature
3.5 Grading system
| Érdemjegy | Pontszám (P) |
| jeles (5) | |
| jó (4) | |
| közepes (3) | |
| elégséges (2) | |
| elégtelen (1) |
3.6 Retake and repeat
3.7 Estimated workload
| Tevékenység | Óra/félév |
| Összesen |
3.8 Effective date
1 September 2022
This Subject Datasheet is valid for:
Inactive courses