Chips for AI

The aim of the microcredential course is to equip learners with systematic knowledge and skills for the development of artificial intelligence chips.

1. Compares CNNs, RNNs, and Transformers and understands their impact at the operator level.
2. Distinguishes between Edge and Cloud AI trade-offs, including latency, privacy, cost, energy consumption, and availability.
3. Applies quantization and pruning techniques to reduce latency, memory usage, and energy consumption while maintaining accuracy.
4. Compares CPUs, GPUs, TPUs, and ASIC accelerators for AI workloads.
5. Explains the trade-offs between FPGA- and ASIC-based solutions for AI acceleration and demonstrates proficiency in FPGA design methodology.
6. Understands the hardware architecture of AI System-on-Chip (SoC) solutions, can use the SystemVerilog hardware description language, and understands the key stages of chip design (synthesis and physical implementation).

Bachelor's degree. Prior knowledge of machine learning and digital electronics design.

Graded

Company experts, engineers and chip designers who develop or select chips for running AI applications.