Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

Internet of Things > Lecture 11 - Tiny Machine Learning > Flashcards

Lecture 11 - Tiny Machine Learning Flashcards

1
Q

What is trustworthy AI?

A

Trustworthy AI should be:
* Lawful - respecting all applicable laws and regulations
* Ethical - respecting ethical principles and values
* Robust - both from a technical perspective while taking into account its social environment ( e.g fairness, inclusivity, alignment with social norms and values etc)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What are the EU Ethics guidelines for trustworthy AI?

A
  • Human agency & oversight
  • Technical robustness & safety
  • Privacy & data governance
  • Transparency
  • Diversity, fairness & non-discrimination
  • Societal & environmental wellbeing
  • Accountability
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What is human agency and oversight?

A
  • AI systems should empower human beings, allowing
    to make informed decisions and fostering their fundamental rights.
  • The allocation of functions between humans and AI systems should follow human-centric design principles and leave meaningful opportunity for human choice.
  • At the same time, proper oversight mechanisms need to be ensured, which can be achieved through human-in-the-loop, human-on-the-loop, and humanin-command approaches.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What is technical robustness and safety?

A
  • AI systems need to be resilient and secure.
  • They need to be safe, ensuring a fall back plan in case something goes wrong.
  • They need to be accurate, reliable and reproducible. That is the only way to ensure that also unintentional harm can be minimized and prevented
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What is privacy and data governance?

A

Besides ensuring full respect for privacy and data protection, adequate data governance mechanisms must also be ensured, taking into account the quality and integrity of the data, and ensuring legitimised access to data.
To allow individuals to trust the data gathering process, it must be ensured that data collected about them will not be used to unlawfully or unfairly discriminate against them.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What is privacy and data governance - Quality and integrity of data?

A

Quality and integrity of the data: When data is gathered, it may contain socially constructed biases, inaccuracies, errors and mistakes. This needs to be addressed prior to training with any given data set.
Access to data:
* Data protocols governing data access should be put in place.
* These protocols should outline who can access data and under which circumstances.
* Only duly qualified personnel with the competence and need to access individual’s data should be allowed to do so.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What is transparency?

A

The data, system and AI business models should be transparent. Traceability mechanisms can help achieving this.
Moreover, AI systems and their decisions should be explained in a manner adapted to the stakeholder concerned. Humans need to be aware that they are interacting with an AI system and must be informed of the system’s capabilities and limitations.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What is diversity, non-discrimination and fairness?

A
  • Unfair bias must be avoided, as it could have multiple negative implications, from the marginalization of vulnerable groups, to the exacerbation of prejudice and discrimination.
  • Fostering diversity, AI systems should be accessible to all, regardless of any disability, and involve relevant stakeholders throughout their entire life circle.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What is societal and environmental well-being?

A

AI systems should benefit all human beings, including future generations.
It must hence be ensured that they are sustainable and environmentally friendly.
Moreover, they should take into account the environment, including other living beings, and their social and societal impact should be carefully considered.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

What is accountability - in reference to AI?

A
  • Mechanisms should be put in place to ensure responsibility and accountability for AI systems and their outcomes.
  • Auditability, which enables the assessment of algorithms, data and design processes plays a key role therein, especially in critical applications.
  • Further, adequate and accessible redress should be ensured.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What are the limitations

A

REFER TO SLIDES FOR LINKS TO STUDY FROM

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

What are some performance metrics for model evaluation?

A
  • Confusion Matrix
  • Precision, Recall, and F1 Score
  • Balanced Accuracy
  • Receiver Operator Characteristics Curve (ROC)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

What is some of the terminology used in pattern classification?

A

Often used in Pattern Classification Problems:
True positive
 The object is there and our classifier says it is there
True negative
 The object is not there and our classifier says it is not there
False negative (false misses)
 The object is there and our classifier says it is not there
False positive (false hits)
 The object is not there and our classifier says it is there

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What is a confusion matrix?

A

REFER TO SLIDES FOR EXAMPLE AND FORMULA

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

What is limitation of accuracy?

A

REFER TO SLIDES FOR EXAMPLE

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What is true postive rate and false positive rate?

A

REFER TO SLDIES FOR FORMULA

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

What is sensitivity and specificity? - GO OVER LECTURE RECORDING FOR THIS

A
  • Sensitivity (True Positive Rate) is the probability of a positive test result, conditioned on the individual truly being positive.
  • Sensitivity Formula: REFER TO SLIDES
  • Specificity (True Negative Rate) is the probability of a negative test result, conditioned on the individual truly being negative.
  • # Specificity Formula: REFER TO SLIDES
  • Sensitivity
     Probability of a true-positive = TP/(TP+FN)
  • Specificity
     Probability of a true-negative = TN/(TN+FP)
  • The probability of a correct decision = (TP+TN)/S,
    where S is the total number of samples
18
Q

What is precision and Recall?

A

Precision is the ratio between true positives versus allpositives
Recall is the measure of how accurate the model is in identifying true positives
REFER TO SLIDES FOR FORMULA FOR BOTH

19
Q

Formula for precision, recall and accuracy

A

Precision = TP/(TP+FP)
Recall = TP/(TP+FN)
Accuracy = (TP+TN)/(TP+TN+FP+FN)

20
Q

What is an F1 score?

A

F1 score takes into account both precision and recall and is based on a balance of the two.

The F1 score is a useful performance metric in machine learning, especially for imbalanced classification problems, where the number of samples in each class is not equal. The F1 score combines two important metrics: precision and recall, into a single value, helping to balance their trade-off.
REFER TO SLIDES FOR FORMULA

21
Q

What is balanced accuracy?

A

Balanced accuracy provides a more insightful measure by accounting for both your model’s sensitivity (true positive rate) and specificity (true negative rate). This makes it particularly valuable in real-world scenarios where imbalanced data is common, and the minority class is usually more important.
* Balanced Accuracy accounts for the performance on both the positive and negative classes, making it particularly useful in imbalanced datasets.
* It is the average of Sensitivity(Recall) and Specificity (True Negative Rate).
Balanced Accuracy = (sensitivity + specificity) / 2
* Standard Accuracy may be misleading in imbalanced datasets because a model could achieve high accuracy simply by predicting the majority class, while balanced accuracy ensures both classes are considered equally.

22
Q

What is parameters vs performance?

A
  • Once we have designed our classifier, we invariably have some parameters we’d like to adjust. e.g. – Prior probability, Threshold
  • The optimal classifier is one with sensitivity (Probability of True Positive) as close to 100% as possible, and at the same time with specificity (Probability of True Negative) as close to 100% as possible
23
Q

What is ROC

A

Developed in 1950s for signal detection theory to analyze noisy signals
– Characterize the trade-off between positive hits and false alarms
ROC curve plots TP (on the y-axis) against FP (on the x-axis)
Performance of each classifier represented as a point on the ROC curve
– changing the threshold of algorithm, sample distribution or cost matrix changes the location of the point.
* A Receiver Operating Characteristic Curve (ROC) is a standard technique for summarizing classifier performance over a range of trade-offs between true positive (TP) and false positive (FP) error rates (Sweets, 1988).
* ROC curve is a plot of sensitivity (the ability of the model to predict an event correctly) versus 1-specificity for the possible cut-off classification probability values .

24
Q

ROC GRAPH

A

REFER TO SLIDES FOR EXAMPLE (SLIDE 101 and 102)

25
Q

What is TinyML?

A

Tiny machine Learning (TinyML) is a fast-growing field of machine learning technologies and applications including algorithms, hardware, and software capable of performing on-device sensor data analytics at extremely low power, typically in the mW range and below, and hence enable a variety of alwayson ML use-case on battery-operated devices.

26
Q

What makes TinyML?

A

A combination of embedded systems and machine learning. Execute the machine learning at the tiny endpoint devices rather than in the
powerful general computer or cloud platform.

27
Q

EXAMPLE OF TINYML - GOOGLE

A

REFER TO SLIDES (FROM SLIDE 27 TO SLIDE 30)

28
Q

Endpoint of sensors

A

REFER TO SLIDES (SLIDE 31)

29
Q

Microcontrollers are (advantages/benefits)?

A

Demanded - high demand
Cheap
Ultra-low power system used
Used frequently, less than 1% of unstructured data is analysed or used at all

30
Q

What are the challanges of TinyML

A

Hardware

31
Q

What is difference between a microprocessor and microcontroller?

A

Microprocessor
* Heart of a computer system
* Just the processor, memory and storage are external
* Mainly used in general purpose systems like laptops, desktops and servers
* Offers flexibility in design
* System size is big

Microcontroller
* Heart of an embedding system
* Memory and storage are all internal to the system
* Mainly used in specialized, fixed function systems like phones, MP3 players, etc.
* Limited flexibility in design
* System size is tiny

REFER TO SLIDES

32
Q

What type of operating system do microprocessors use?

A

arm MBED OS
free RTOS
Embedded sys.
- Mbed is a development platform and operating system fodevices based on 32-bit ARM Cortex-M microcontrollers.

33
Q

What is real time OS?

A
  • A real-time operating system (RTOS) is an operating system (OS) for real-time computing applications that processes data and events that have critically defined time constraints.
  • An RTOS is distinct from a time-sharing operating system that manages the sharing of system resources with a scheduler, data buffers, or fixed task prioritization in a multitasking or multiprogramming environments.
  • All processing must occur within the defined constraints.
  • Event-driven systems switch between tasks based on their priorities, while time-sharing systems switch the task based on clock interrupts
34
Q

Comparison of Embedded system

A

REFER TO SLIDES (60)

35
Q

Microcontroller usage - deterministic vs problematic

A

REFER TO SLIDES (61)

36
Q

Libraries and portability

A

REFER TO SLIDES (62+)

37
Q

Embedded hardware vs embedded software?

A

Embedded hardware: is extremely limited in performance, power consumption and storage
Embedded software: is not as portable and flexible as mainstream computing

38
Q

Machine Learning Evolution

A

REFER TO SLIDES (72+)

39
Q

What are some model compression techniques - pruning?

A
  • Pruning: Pruning reduces the complexity of a neural network by identifying and eliminating weights or neurons that contribute little to the model’s final predictions. This process leads to smaller, faster models with reduced memory and computational requirements
40
Q

What are some model compression techniques - quantization

A

Mapping of infinite values to a smaller set of finite values

41
Q

What are some model compression techniques - knowledge distilation

A

The main goal of knowledge distillation is to transfer the knowledge learned by a large model into a smaller model, enabling faster inference and reduced memory requirements while maintaining comparable performance.

42
Q

What are some tools for model compression?

A

TensorFlow Lite
arm
STM32 Cube.AI
Edge Impulse
MICROEJ
NVIDIA Jetson

Internet of Things (18 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2024 Bold Learning Solutions. Terms and Conditions