EE543 NEUROCOMPUTERS AND DEEP LEARNING
| Course Code: | 5670543 |
| METU Credit (Theoretical-Laboratory hours/week): | 3 (3.00 - 0.00) |
| ECTS Credit: | 8.0 |
| Department: | Electrical and Electronics Engineering |
| Language of Instruction: | English |
| Level of Study: | Graduate |
| Course Coordinator: | Prof.Dr. UĞUR HALICI |
| Offered Semester: | Fall Semesters. |
Course Objectives
This course aims to give an insight on various aspects of neurocomputers with an emphasize on deep learning approaches:
- Detailed knowledge on deterministic and probabilistic neurons, feed forward and recurrent neural networks, deep structures of feed forward and recurrent neural networks,
- Detailed knowledge various deep learning algorithms and applying this knowledge for solving practical problems,
Course Content
Various aspects of neurocomputers emphasizing deep learning approaches. Brainlike computing, characteristics of neurocomputers. Deterministic, probabilistic and spiking neuron models. Feed forward and recurrent neural networks. Deep structures of feed forward and recurrent neural networks. Deep learning: supervised and unsupervised machine learning algorithms for various deep neural networks. Neuromorphic chips.
Course Learning Outcomes
Student, who passed the course satisfactorily will be able to:
- Describe brain-like computing and differences of neurocomputers from conventional digital computers,
- Understand various aspects of neurocomputers such as deterministic, probabilistic and spiking neurons, feed forward and recurrent neural networks, deep structures of feed forward and recurrent neural networks, deep learning algorithms for training them.
- Applying deep learning algorithms to solve realistic problems and measure their performances.