Course detail
Convolutional Neural Networks
FIT-KNNAcad. year: 2019/2020
Solutions based on machine learning approaches gradually replace more and more hand-designed solutions in many areas of software development, especially in perceptual task focused on information extraction from unstructured sources like cameras and microphones. Today, the dominant method in machine learning is neural networks and their convolutional variants. These approaches are at the core of many commercially successful applications and they push forward the frontiers of artificial intelligence.
Language of instruction
Number of ECTS credits
Mode of study
Guarantor
Learning outcomes of the course unit
Students will acquire team work experience during project work and they will acquire basic knowledge of python libraries for linear algebra and machine learning.
Prerequisites
Co-requisites
Planned learning activities and teaching methods
Assesment methods and criteria linked to learning outcomes
- Project concluded by public presentation - 65 points.
- Three tests during the semester - 35 points.
Exam prerequisites:
Acquiring at least 50 points.
Course curriculum
Work placements
Aims
Specification of controlled education, way of implementation and compensation for absences
Recommended optional programme components
Prerequisites and corequisites
Basic literature
Recommended reading
Goodfellow, I., Bengio, Y., Courville, A.: Deep Learning. MIT Press, 2016.
Li, Fei-Fei, et al.: CS231n: Convolutional Neural Networks for Visual Recognition. Stanford, 2018.
Classification of course in study plans
- Programme MITAI Master's
specialization NBIO , 0 year of study, summer semester, compulsory
specialization NSEN , 0 year of study, summer semester, elective
specialization NVIZ , 0 year of study, summer semester, compulsory
specialization NGRI , 0 year of study, summer semester, elective
specialization NISD , 0 year of study, summer semester, elective
specialization NSEC , 0 year of study, summer semester, elective
specialization NCPS , 0 year of study, summer semester, elective
specialization NHPC , 0 year of study, summer semester, elective
specialization NNET , 0 year of study, summer semester, elective
specialization NMAL , 0 year of study, summer semester, compulsory
specialization NVER , 0 year of study, summer semester, elective
specialization NIDE , 0 year of study, summer semester, elective
specialization NEMB , 0 year of study, summer semester, elective
specialization NSPE , 0 year of study, summer semester, compulsory
specialization NADE , 0 year of study, summer semester, elective
specialization NMAT , 0 year of study, summer semester, elective
specialization NISY , 0 year of study, summer semester, elective
Type of course unit
Lecture
Teacher / Lecturer
Syllabus
- Introduction, linear models. loss function, learning algorithms and evaluation. (Organization, basics of neural networks, numpy tutorial)
- Fully connected networks, loss functions for classification and regression. (PyTorch simple regression, Image Classification)
- Convolutional networks, locality in equivariance of computation. (Lecture)
- Generalization, regularization, data augmentation. multi-task learning and pre-training. (Optimizers)
- Problems during network optimization and batch normalization. Existing image classification network architectures. (Image classification networks)
- Object detection: MTCNN face detector, R-CNN, Fast R-CNN, Faster R-CNN, YOLO, SSD.
- Semantic and instance segmentation. Connections to estimation of depth, surface normals, shading and motion.
- Learning similarity and embedding. Person identification.
- Recurrent networks and sequence processing (text and speech). Connectionist Temporal Classification (CTC). Attention networks.
- Language models. Basic image captioning networks, question answering and language translation.
- Generative models. Autoregressive factorization. Generative Adversarial Networks (GAN, DCGAN, cycle GAN).
- Reinforcement learning. Deep Q-network (DQN) and policy gradients.
- Overview of emerging applications and cutting edge research.
Project
Teacher / Lecturer
Syllabus
Individual assignments - proposed by students, approved by the teacher. Components:
- Problem Formulation, team formation.
- Research of existing solutions and usefull tools.
- Baseline solution and evaluation proposal.
- Data collection.
- Experiments, testing and gradual improvement.
- Final report and Presentation of the project.