IN3310 – Deep Learning for Image Analysis

Schedule, syllabus and examination date

Course content

This course teaches common methods in deep learning applied to image data, covering key deep learning algorithms and concepts for training neural networks. The course's focus is on supervised learning and image classification. Nevertheless, the course will also introduce other common learning regimes and image analysis tasks, such as image segmentation and object detection.

Learning outcome

After this course you will:?

  • understand how neural networks are built and how backpropagation works;?
  • understand empirical risk minimization and be familiar with popular loss and activation functions;
  • understand key mathematical insights and intuition behind the training process, how to handle distribution shifts and generalize, and regularization techniques;
  • know how to train a neural network from scratch and basics of popular first-order optimization methods, use pre-trained models, fine tune the neural networks, and discern when to use each approach to solve a problem;
  • understand how to treat the data (augment and clean it) to improve the efficacy of neural networks;?
  • know different network architectures and in what contexts they are suitable;?
  • understand the inductive bias of locality imposed by convolutions, its implementation on convolutional neural networks, and its application to imaging;?
  • understand the supervised learning regime;?
  • know how to apply deep learning to solve problems that depend on imaging data, for example, image classification, object segmentation, object detection, among others; ?
  • have experience in using Pytorch.?

Admission to the course

Students who are admitted to study programmes at UiO must each semester register which courses and exams they wish to sign up for?in Studentweb.

Special admission requirements

In addition to fulfilling the?Higher Education Entrance Qualification, applicants have to meet the following special admission requirements:

  • Mathematics R1 or Mathematics (S1+S2)

The special admission requirements may also be covered by equivalent studies from Norwegian upper secondary school or by other equivalent studies. Read more about?special admission requirements?(in Norwegian).

Formal prerequisite knowledge

FYS-STK3155 – Applied Data Analysis and Machine Learning or STK2100 – Machine Learning and Statistical Methods for Prediction and Classification?

The student should have a strong background in?

  • programming, for this we recommend IN2010;?
  • mathematics (calculus and linear algebra), we recommend that the student has taken MAT1110, and it will be good to have MAT1120;?

Its also recommended to have some knowledge on applications related to images, such as from IN2070.?

Overlapping courses

Teaching

2 hours of lectures and 2 hours of exercises each week.

This course has mandatory assignments that must be approved before the exam.?Read more about requirements for submission of assignments, group work and legal cooperation under guidelines for mandatory assignments.

Examination

Written exam (4 hours).?

All mandatory assignments must be approved before you can take the exam.?

It will also be counted as one of your three attempts to sit the exam for this course, if you sit the exam for one of the following courses: IN4310 – Deep Learning for Image Analysis, IN5400 – Machine Learning for Image Analysis (continued), IN9400 – Machine Learning for Image Analysis (discontinued). INF5860, INF9860

Examination support material

No examination support material is allowed.?

Grading scale

Grades are awarded on a scale from A to F, where A is the best grade and F is a fail. Read more about the grading system.?

More about examinations at UiO

You will find further guides and resources at the web page on examinations at UiO.

Last updated from FS (Common Student System) Dec. 22, 2024 7:02:07 AM

Facts about this course

Level
Bachelor
Credits
10
Teaching
Spring
Examination
Spring
Teaching language
English