7. EDGE AI TOOLS & CHALLENGES

Question 1

There are a plenty of benchmark datasets for deep learning which are targeted at edge devices
a. True
b. False

Answer 1

b. False

Question 2

The aggregation of dataset and models in an Edge AI environment is known as Edge ______

Answer 2

Caching

Question 3

Which of the following software development kit is dedicated to STM32 microcontrollers?
a. Nvidia TensorRT
b. AMD DNNDK
c. None of the following
d. X-Cube AI

Answer 3

d. X-Cube AI

Question 3

What is Edge Fleet Management?
a. None of the following
b. It is a hub where all the edge devices will publish/subscribe the data
c. It is a network of edge devices
d. It is a process of compressing DL algorithms to be run on an Edge Device

Answer 3

b. It is a hub where all the edge devices will publish/subscribe the data

Question 4

Which of the following languages/software is preferred for implementing on an edge device?
a. Python
b. Any of the following
c. C or C++
d. Matlab

Answer 4

c. C or C++

Question 5

Which of the following is a software tool for edge DL inference?
a. ExecuTorch
b. All of them
c. Tensorflowlite
d. Onnx

Answer 5

b. All of them

Question 6

Which of the following conditions is not important for an Edge AI deployment?
a. Simultaneous design of algorithms with hardware
b. All of the following
c. The environmental conditions of edge AI deployment doesn’t really matter
d. Tools for automatic mapping of trained models

Answer 6

c. The environmental conditions of edge AI deployment doesn’t really matter

Question 7

Which software tool supports the Raspberry Pi for edge inference?
a. Embedded Learning Library (ELL)
b. Intel OpenVINO Toolkit
c. None of the following
d. Matlab Deep Learning HDL Toolbox

Answer 7

a. Embedded Learning Library (ELL)

Question 8

The process of identifying the memory requirements and latency for running an ML algorithm on a specific hardware is known as?

Answer 8

Profiling

Question 9

It is not feasible to implement an unsupervised ML algorithm like k-Nearest Neighborhood (k-NN) on an edge device.
a. True
b. False

Answer 9

b. False

Question 10

It is possible to run an ML algorithm on a field programmable gate device (FPGA) or an application specific integrated circuit (ASIC)
a. True
b. False

Answer 10

a. True

Question 11

What is the trimmed version of Python 3.x which could be run on an edge device?

Answer 11

MicroPython

Question 12

There are plenty of software frameworks available for neural architecture searches for edge devices.
a. True
b. False

Answer 12

b. False

Question 13

In Hierarchical Federated Learning, which of the following is correct?
a. The model is trained at the edge
b. The model is trained in the cloud
c. The model is trained in the cloud and at the edge
d. None of the above

Answer 13

c. The model is trained in the cloud and at the edge

Question 14

When the dataset is not shared, but the model is locally trained and shared across multiple compute devices, the framework is known as
a. Localized Learning
b. Federated Learning
c. Globalized Learning
d. Edge Learning

Answer 14

b. Federated Learning