Compute Platforms for AI and Embedded Processing (B-KUL-H05H2A)
Aims
The aim of this course is to learn to design larger digital integrated systems-on-chip (SoCs), and this especially in the area of AI and perceptive sensory systems. The course will start from the design, computer architecture and methodology knowledge gathered from previous courses. Special attention will be given to :
acquiring knowledge on state-of-the-art efficiency enhancement techniques used in processors and accelerators
the design trade-offs between performance, cost, power usage and flexibility
- the tight interplay and co-optimizations between hardware and algorithms (HW-alg co-design).
Previous knowledge
A succesful exam of the following courses is advised
- h09j6 design of digital platforms
- h05d3a computer architectures
Order of Enrolment
Mixed prerequisite:
You may only take this course if you comply with the prerequisites. Prerequisites can be strict or flexible, or can imply simultaneity. A degree level can be also be a prerequisite.
Explanation:
STRICT: You may only take this course if you have passed or applied tolerance for the courses for which this condition is set.
FLEXIBLE: You may only take this course if you have previously taken the courses for which this condition is set.
SIMULTANEOUS: You may only take this course if you also take the courses for which this condition is set (or have taken them previously).
DEGREE: You may only take this course if you have obtained this degree level.
(FLEXIBLE (H05D3A) OR FLEXIBLE (H05D5A))
The codes of the course units mentioned above correspond to the following course descriptions:
H05D3A : Computer Architectures
H05D5A : Computerarchitecturen
Is included in these courses of study
- Master in de nanowetenschappen, nanotechnologie en nano-engineering (Leuven) 120 ects.
- Master of Nanoscience, Nanotechnology and Nanoengineering (Leuven) 120 ects.
- Courses for Exchange Students Faculty of Engineering Science (Leuven)
- Master in de ingenieurswetenschappen: elektrotechniek (Leuven) (Elektronica en chipontwerp) 120 ects.
- Master in de ingenieurswetenschappen: elektrotechniek (Leuven) (Informatiesystemen en signaalverwerking) 120 ects.
- Master of Electrical Engineering (Leuven) (Electronics and Chip Design) 120 ects.
- Master of Electrical Engineering (Leuven) (Information Systems and Signal Processing) 120 ects.
Activities
2.41 ects. Compute Platforms for AI and Embedded Processing: Lecture (B-KUL-H05H2a)
Content
This course will dive deeper into the following topics:
- Evolution in (embedded) processor chips
- Overview of AI hardware, computer architectures for AI and deep learning
- Processor tecniques for power management and adaptive operation of large SoC
- Design verficiation using system verilog
- Guest lecture on industrial SoC development
Course material
Study cost: 1-10 euros (The information about the study costs as stated here gives an indication and only represents the costs for purchasing new materials. There might be some electronic or second-hand copies available as well. You can use LIMO to check whether the textbook is available in the library. Any potential printing costs and optional course material are not included in this price.)
Ad hoc material put on Toledo.
Format: more information
There are theoretical classes and exercise sessions.
0.59 ects. Compute Platforms for AI and Embedded Processing: Exercises and Laboratory Sessions (B-KUL-H05H3a)
Content
This class encompasses a mixture of prerecorded knowedge clips, more interactive class sessions and design exercises.
Course material
Ad hoc material put on Toledo.
Evaluation
Evaluation: Compute Platforms for AI and Embedded Processing (B-KUL-H25H2a)
Explanation
There are three types of questions:
- questions about the theory taught in the classes
- questions about a related paper
- questions about the design exercise, which needs to be submitted at the end of the project (deadline announced in the work sessions).
Information about retaking exams
Students taking this exam in the 3rd examation period, will get a modified design exercise, and need to submit this again before the exam. It will again be orally discussed at the exam.