Z-ENG: Evaluation of Neural Network Execution on Google Edge TPU System

Félév: 2024-2025 ősz

Kategória: Nincs megadva

Téma leírása

Machine learning applications have two main lifecycles: the learning phase, where the neural network is optimized and tuned for a given task, and the inference phase, where the trained network is used to solve the problem. Machine learning typically takes place on high-performance computers equipped with GPUs during the learning phase, while inference usually occurs on low-power, inexpensive embedded systems with low computational capacity. These systems are often equipped with specialized neural network accelerator modules.

Different accelerators have different hardware architectures, resulting in varying characteristics and performance. Specific layers and operations of the neural network can be executed with varying efficiency by different accelerators, making different execution units ideal choices for different network architectures.

The student's task is to examine the computational capabilities of a specific accelerator, the Google Edge TPU (Tensor Processing Unit), across a wide range of convolutional neural network architectures. Using the measurement results, the student will propose recommendations for optimizing the efficiency of neural network execution on the TPU with respect to network architecture. As part of the task, the student will become familiar with the characteristics of convolutional neural networks, Google Edge TPU, and the TensorFlow framework.

References:
Google Edge TPU: https://cloud.google.com/edge-tpu/
TensorFlow Lite: https://www.tensorflow.org/lite

To solve this task, the student will receive assistance from the employees of the Continental AI Development Center.

If you are interested in the topic, be sure to contact Dávid Sik by email before applying, indicating the selected topic, training level, major and the planned project subject.

Külső partner: Continental Autonomous Mobility Hungary

Maximális létszám: 1 fő