INTD 161 Pt 3

Choose a study mode

Play Quiz
Study Flashcards
Spaced Repetition
Chat to Lesson

Podcast

Play an AI-generated podcast conversation about this lesson
Download our mobile app to listen on the go
Get App

Questions and Answers

How can AI encode bias, as exemplified by the Dall-E image generation model?

  • By reinforcing stereotypes present in the training data. (correct)
  • By randomly generating images without any specific pattern.
  • By prioritizing creative and imaginative content over realistic portrayals.
  • By accurately reflecting the diversity present in society.

What is the primary distinction between AI and ML based on the lecture?

  • ML involves systems with goals that make decisions, while AI focuses on data extraction.
  • ML extracts knowledge from data to build models, while AI encompasses any system that mimics intelligence. (correct)
  • AI is a subset of ML focused on complex problem-solving, whereas ML is a broader field.
  • AI extracts knowledge from data to build models, while ML mimics intelligence directly.

Which of the following best describes the role of 'labels' in machine learning?

  • They outline the ethical guidelines for the model’s application.
  • They are used to define the algorithms to be used in the model.
  • They provide a description of the data collection process.
  • They implicitly tell the model what question to answer and what the correct answers are. (correct)

What is the role of data in machine learning models, according to the lecture?

<p>To train models and distill information and knowledge. (B)</p> Signup and view all the answers

An ML model is presented with an image of an animal and outputs 'dog'. What type of ML model is it, and what kind of label does it use?

<p>Classification model; category label. (D)</p> Signup and view all the answers

In the context of machine learning, what does it mean for a model to 'generalize' well?

<p>The model can make accurate predictions on new, unseen data. (C)</p> Signup and view all the answers

Which statement accurately describes the utilization of decision trees in machine learning?

<p>They offer a visually intuitive representation of decision-making processes. (C)</p> Signup and view all the answers

What is a primary advantage of using decision trees in machine learning?

<p>Their great explainability. (A)</p> Signup and view all the answers

What does the 'Bag of Words' representation primarily aim to achieve in natural language processing?

<p>Converting text into a numerical format suitable for machine learning models. (B)</p> Signup and view all the answers

In the context of count-based models in NLP, what does the term 'near' typically refer to when incrementing the count in each cell?

<p>Words that appear in close proximity to each other, like within the same sentence or document. (B)</p> Signup and view all the answers

What is the advantage of updating ML models versus expert systems when inaccuracies are found?

<p>ML models can be updated more easily, allowing for quicker correction of inaccuracies. (C)</p> Signup and view all the answers

Which of the following is a task that can be approached using supervised learning?

<p>Developing a model that predicts housing prices based on features of the property. (A)</p> Signup and view all the answers

How do Large Language Models (LLMs) predict the next word in a sentence?

<p>By analyzing a context window of words to predict the subsequent word. (D)</p> Signup and view all the answers

In the context of supervised learning, what is the source of the data's labels used to train the models?

<p>Provided by human experts or annotators. (A)</p> Signup and view all the answers

Why is it important to be careful about bias in data used for AI systems?

<p>AI systems can encode and perpetuate biases present in the data, leading to unfair or discriminatory outcomes. (D)</p> Signup and view all the answers

Which of the following tasks is best best described as AI but not ML?

<p>Expert system MYCIN. (C)</p> Signup and view all the answers

Which of the following tasks is best described as both AI and ML?

<p>Spam filter. (D)</p> Signup and view all the answers

Which of the following best describes the relationship between data and ML model?

<p>Data is the first building block to ML models to extract information. (C)</p> Signup and view all the answers

How do you input a picture into an ML model and what type of model does it use?

<p>Pictures are inputed as numbers into ML model. (A)</p> Signup and view all the answers

What is the purpose of labels in ML model?

<p>Labels is for the model to know what question we want to be answered and that the right answers are. (D)</p> Signup and view all the answers

What is the difference(s) between Classification and Regression?

<p>Classification is the output of category output but Regression is the output of continues output. (A)</p> Signup and view all the answers

What is unsupervised learning?

<p>It focuses on scenarios when the model does not learn from labeled data. (C)</p> Signup and view all the answers

What did LLMs learn?

<p>That &quot;the and animal&quot; are import words in a sentence. (A)</p> Signup and view all the answers

According to the video, what is one of the most basic paradigms in ML?

<p>Supervised learning. (C)</p> Signup and view all the answers

ML models are better at the examples we give it, what is the name of it?

<p>data. (D)</p> Signup and view all the answers

Decision trees are what type of machine to use to separate the data?

<p>Observations, given precipitations, clothes, and other environment labels. (B)</p> Signup and view all the answers

What will happen if there is more data to create ML models to do a prediction?

<p>ML models provide generalize to new data. (D)</p> Signup and view all the answers

What happened if the temperature is -5, with Snow, on Wednesday wearing casual, what will the label show?

<p>The label should show Bus because Wednesday does to generalize the temp and precipitations. (A)</p> Signup and view all the answers

What are decision trees?

<p>Decision tree is a type of machine learning algorithms. (B)</p> Signup and view all the answers

Which of the following best describes the AI system MYCIN?

<p>It is not an ML model but an expert rule-based system. (C)</p> Signup and view all the answers

What is needed when the AI system is inaccurate?

<p>We can update the knowledge in ML models. (D)</p> Signup and view all the answers

The weather man forecast the weather as raining with 100 percentage and thunderstorm for Tuesday at 10:00AM, what are the category and values?

<p>Category(Rainy for thunderstorm) and values(percentage of temp, humidity). (C)</p> Signup and view all the answers

Which statement is true regarding data and ML?

<p>ML is always dependent to data. (A)</p> Signup and view all the answers

What does count-based models mean for NLP?

<p>Conversion of each word to a bunch of numbers, based on its relations to other words. (A)</p> Signup and view all the answers

If you provide a cat list of pictures and a dog list of pictures. Then the ML model shows a dinosaur. What do you call that?

<p>ML Models learn about all the questions in the photos and can answer all the results. (A)</p> Signup and view all the answers

What aspect of AI systems requires careful attention to prevent biased outputs?

<p>The data they are trained on. (C)</p> Signup and view all the answers

What term describes the concept of data in machine learning?

<p>Information. (B)</p> Signup and view all the answers

In the machine learning pipeline, what is the purpose of the 'Training' stage?

<p>To extract info/knowledge from data to build the model (D)</p> Signup and view all the answers

In a machine learning context, what is the role of 'labels'?

<p>Indicate the correct answer/output/prediction for a given input (A)</p> Signup and view all the answers

What is the primary objective of converting text into numerical representations in NLP?

<p>To enable mathematical calculations for machine learning models. (B)</p> Signup and view all the answers

In count-based models for NLP, what does incrementing the count in a cell signify?

<p>The co-occurrence of two words within a defined context. (A)</p> Signup and view all the answers

What is a key characteristic of supervised learning?

<p>Models learn from labeled datasets to imitate provided answers. (C)</p> Signup and view all the answers

What is the practical implication of the statement that an ML model is an 'imitation machine'?

<p>The model's accuracy is limited by the quality and nature of the training data. (A)</p> Signup and view all the answers

What is the core difference between how an expert system and a supervised learning system provide answers?

<p>Expert systems answer based on predefined rules, while supervised learning systems attempt to imitate answers provided in training data. (D)</p> Signup and view all the answers

Why are ML models easier to update when inaccuracies are found, compared to expert systems?

<p>ML models can be retrained with new data to adjust their behavior. (B)</p> Signup and view all the answers

What are the main components to consider when determining the type of transportation a person is using?

<p>All the above. (B)</p> Signup and view all the answers

What does it mean for a machine learning model to 'generalize' well?

<p>It can accurately make predictions on new, unseen data. (B)</p> Signup and view all the answers

In the context of the YouTube video recommendation system, what serves as the 'input' for the ML model?

<p>The user's viewing history, current video, likes, dislikes, etc. (B)</p> Signup and view all the answers

In Google Lens, what is the primary role of machine learning?

<p>To analyze the objects in order to characterize them. (C)</p> Signup and view all the answers

What best describes the role of data in ChatGPT?

<p>It is used to train the model on language patterns and knowledge. (C)</p> Signup and view all the answers

Consider the statement: 'I had to take my _____ to the vet.' What concept does this illustrate in language models?

<p>The importance of context in determining word choice. (D)</p> Signup and view all the answers

In Large Language Models (LLMs), what is the purpose of the 'context window'?

<p>To focus on a subset of text for predicting the next word. (C)</p> Signup and view all the answers

Given the data of temperature, precipitation, day and clothes; What type of ML is used to produce the output (label)?

<p>Supervised Learning (B)</p> Signup and view all the answers

How would you describe passive imitation?

<p>Model has no agency: it learns passively (B)</p> Signup and view all the answers

What does it mean for ML to be great explainability?

<p>Easy for us to understand how and why a decision tree makes a certain prediction (D)</p> Signup and view all the answers

When should we be cautious of bias when training data?

<p>If AI systems can encode bias if not well taken care of! (B)</p> Signup and view all the answers

When is assignment #2 due?

<p>Next Tuesday! (A)</p> Signup and view all the answers

What does Building blocks of ML consists of?

<p>All the above (B)</p> Signup and view all the answers

Why does 3 submissions exist for assignment #2?

<p>All the above (D)</p> Signup and view all the answers

Where did the labels/answers come mostly from in Supervised Learning?

<p>People (C)</p> Signup and view all the answers

Which of these applications of AI is also ML?

<p>Spam filter (D)</p> Signup and view all the answers

What does Data consists of?

<p>A collection of discrete or continuous values that convey Information (B)</p> Signup and view all the answers

How to you input a picture into an ML model?

<p>Convert it into numbers (B)</p> Signup and view all the answers

What does classify do regarding the types of news article?

<p>Many options (C)</p> Signup and view all the answers

What are the amount of options when classifying a review type of postive or negative?

<p>Two options (B)</p> Signup and view all the answers

What does 'Large' stand for in LLM?

<p>Huge Neural Nets with Billions of connections &amp; neurons (D)</p> Signup and view all the answers

In Count Based Models, what does 'near' mean?

<p>All the above (B)</p> Signup and view all the answers

Which of this option will the statement refer to 'Cats will Escape'?

<p>Will (A)</p> Signup and view all the answers

Given the temp is high, predict the label. Temp > 0

<p>Walk (C)</p> Signup and view all the answers

What does Generalize mean related to ML?

<p>Generalization is the ability to make predictions on new, unseen data (D)</p> Signup and view all the answers

What is a primary limitation of a perceptron?

<p>Inability to model complex non-linear relationships. (D)</p> Signup and view all the answers

What distinguishes generative learning from supervised learning?

<p>Generative learning uses labels to predict data, while supervised learning predicts labels from data. (A)</p> Signup and view all the answers

Which factor does not significantly influence the decision-making when choosing between different ML models?

<p>The size of the team working on the project. (B)</p> Signup and view all the answers

What is the primary advantage of using generative learning with unlabeled data?

<p>It allows for scaling to large datasets, thus reducing human effort in labeling. (B)</p> Signup and view all the answers

In the context of Large Language Models (LLMs), what is the purpose of the 'context window'?

<p>To provide the model with a subset of text that dictates the next word in a sequence. (B)</p> Signup and view all the answers

What characteristic defines sequence generation models, such as those used for video generation?

<p>They produce data by generating a sequence of data, building upon the previously provided data. (B)</p> Signup and view all the answers

In Generative Adversarial Networks (GANs), what is the role of the 'discriminator'?

<p>To decide if a given output is real or artificial. (A)</p> Signup and view all the answers

Why is it important for generative models to have a large amount of data available?

<p>To create outputs to be more like human outputs. (B)</p> Signup and view all the answers

How do decision trees classify?

<p>By separating data using multiple straight line segments. (A)</p> Signup and view all the answers

What is the goal of the generator in Generative Adversarial Networks (GANs)?

<p>Create training images to make the discriminator make an error. (C)</p> Signup and view all the answers

What best describes the modelling power of the Artificial Neural Network?

<p>Excellent (B)</p> Signup and view all the answers

What could generative models help application(s) related to?

<p>All of the above (D)</p> Signup and view all the answers

In the three models; Perceptron, Decision Tree, and Artificial Neural Network; Which of these models has the simplest interperability?

<p>Perceptron (B)</p> Signup and view all the answers

What does a LLM (Large language Model) look at to predict the next word?

<p>Looks at a context window (B)</p> Signup and view all the answers

Which action does unsupervised learning perform?

<p>None of the above (D)</p> Signup and view all the answers

Why is the output of a generative learning randomized?

<p>To create different predictions (D)</p> Signup and view all the answers

What is meant by needing more than lines?

<p>The strength of ANNs has complex non-linear functions. (A)</p> Signup and view all the answers

With the 3 models; Perceptron, Decision Tree, and ANN. If the interpretability is poor, what model will that be?

<p>ANN (B)</p> Signup and view all the answers

Generative learning commonly has...

<p>No labels (A)</p> Signup and view all the answers

When would a statement like, 'This is a big deal!' be said?

<p>If the model is designed to predict-the-next-word like an LLM. (B)</p> Signup and view all the answers

What is the risk of training the models recursively on data that has been generated?

<p>The models may begin to lose training data (A)</p> Signup and view all the answers

What functions are ANNs used for?

<p>Complex Non-Linear Functions (B)</p> Signup and view all the answers

Is it possible to turn our simple decision tree into a 2d plot?

<p>TRUE (D)</p> Signup and view all the answers

What two networks are used in GAN?

<p>Two artificial networks used to generate interesting novel training data. (A)</p> Signup and view all the answers

Minksky's famous book did what to perceptrons?

<p>Killed perceptrons (D)</p> Signup and view all the answers

Why do we want the model to be creative with a dice?

<p>Both A and C (C)</p> Signup and view all the answers

What does Generative Learning need to generate new images?

<p>Dataset of images (D)</p> Signup and view all the answers

What should happen that tells us to stop?

<p>If generator training goes well, the discriminator gets worse at telling the difference between real and fake. it starts to classify fake data as real, and its accuracy decreases (A)</p> Signup and view all the answers

What is the difference between supervised learning and generative learning?

<p>All of the above (D)</p> Signup and view all the answers

How does Chat GPT generate?

<p>Human-like conversation responses (B)</p> Signup and view all the answers

If chatGPT outputs something that is wrong, will we need to update the answers?

<p>No, Human labels will be needed (B)</p> Signup and view all the answers

Which model can learn complex non-linear functions?

<p>Artificial Neural Networks (A)</p> Signup and view all the answers

What is a characteristic of sequence generation models?

<p>The ability to generate a frame at a time. (C)</p> Signup and view all the answers

What is the purpose of Data?

<p>The information used to Train the model (C)</p> Signup and view all the answers

Is AI in creative activities?

<p>Yes, to write stories articles (B)</p> Signup and view all the answers

Which of this option is right?

<p>All of the above. (D)</p> Signup and view all the answers

What is a key characteristic that distinguishes generative learning from supervised learning?

<p>Generative learning can work with unlabeled data to generate new data, while supervised learning requires labeled data to learn a mapping. (C)</p> Signup and view all the answers

What is the primary goal of the generator network in a Generative Adversarial Network (GAN)?

<p>To create synthetic images that are indistinguishable from real images in the training set. (D)</p> Signup and view all the answers

Why is it often necessary for generative models to be trained with a large amount of data?

<p>To capture the underlying patterns and distribution of the data, enabling the generation of realistic and diverse outputs. (D)</p> Signup and view all the answers

In the context of Large Language Models (LLMs), like ChatGPT, what is the role of the 'context window'?

<p>To serve as the input the LLM looks at to predict the next word in a sequence. (C)</p> Signup and view all the answers

What is the main purpose of the discriminator in a Generative Adversarial Network (GAN)?

<p>To distinguish between real and fake data samples. (A)</p> Signup and view all the answers

What is a key limitation of a perceptron in machine learning?

<p>Its inability to learn non-linear relationships. (D)</p> Signup and view all the answers

What is meant by needing more than lines in machine learning?

<p>Employing complex models for non-linear relationships. (A)</p> Signup and view all the answers

What is the key factor that sets generative learning apart, making it scalable and useful?

<p>Its ability to function effectively with large amounts of unlabeled data, such as for next word prediction. (A)</p> Signup and view all the answers

What prompted skepticism and a subsequent decline in the popularity of research on perceptrons in the late 1960s?

<p>Discovery of their inability to solve problems with non-linear relationships, as highlighted by the XOR problem. (B)</p> Signup and view all the answers

In what context might the statement 'This is a big deal!' be said?

<p>To express the significance of leveraging unlabeled data and scaling learning effectively. (D)</p> Signup and view all the answers

What are Artificial Neural Networks (ANNs) most well known for?

<p>They learn complex non-linear function. (B)</p> Signup and view all the answers

What is the significance of the statement 'Sutton's Bitter Lesson'?

<p>Exploiting computation and data usually leads to the best outcome. (B)</p> Signup and view all the answers

If the interpretability score is poor, what model will that be?

<p>ANN. (B)</p> Signup and view all the answers

Can you always turn our simple decision tree into a 2d plot?

<p>Yes, you can always turn our simple decision tree into a 2d plot. (D)</p> Signup and view all the answers

Why is the output of a generative learning model randomized?

<p>To make results appear more natural and creative. (B)</p> Signup and view all the answers

If it wasn't possible to separate XOR function, then what happened?

<p>First AI winter. (C)</p> Signup and view all the answers

What are two networks that are used in GAN?

<p>Generator, Discriminator. (D)</p> Signup and view all the answers

How and why do you stop a generative learning?

<p>Both A and D. (E)</p> Signup and view all the answers

Why is the Generator helpful to use?

<p>The generator is the useful artifact to the user. (B)</p> Signup and view all the answers

If a Cat picture is given to the model and cat picture is generated, what is this?

<p>Generative Learning. (C)</p> Signup and view all the answers

A pathology foundation model is what type of example?

<p>AI diagnostic in health care. (D)</p> Signup and view all the answers

ChatGPT prompt: Tell me a joke that I can use in a class that teaches Al to the public. ChatGPT generates a joke. What is it?

<p>Why was the computer cold? (A)</p> Signup and view all the answers

What is an accurate description regarding the model's output in generative learning systems?

<p>Often randomized to foster creativity. (B)</p> Signup and view all the answers

What action does unsupervised learning perform on the training dataset?

<p>None of the above. (D)</p> Signup and view all the answers

What does a Large Language Model look at to predict the next word?

<p>It looks at a context window. (B)</p> Signup and view all the answers

What should be considered regarding the topic of Recursive Training?

<p>Can be potentially biased. (B)</p> Signup and view all the answers

When considering bias with generative AI, what action should a user take?

<p>Be aware of the bias generative AI. (C)</p> Signup and view all the answers

Flashcards

What is AI?

Anything that mimics intelligence

What is Machine Learning (ML)?

A subset of AI focusing on extracting knowledge from data to build models for predictions

What is bias in ML?

ML models may encode societal biases if not carefully addressed.

What is Data?

Values that convey information, meaning, or statistics.

Signup and view all the flashcards

What is ML Training?

Extracting info/knowledge from data sets to build a predictive model.

Signup and view all the flashcards

What is ML Prediction?

Using a model to provide an answer to a query.

Signup and view all the flashcards

ML labels

Implicitly tell the model what question we want it to answer and what the right answers are.

Signup and view all the flashcards

ML Classification

The output of the model is a category of the input.

Signup and view all the flashcards

ML Regression

The output of the model is a continuous numerical value.

Signup and view all the flashcards

ML sequence

A sequence of words as output of a model.

Signup and view all the flashcards

Bag of Words

Converting text to numbers so models can do calculations.

Signup and view all the flashcards

Count Based Models

Encoding words based on relations.

Signup and view all the flashcards

Generalization

The ability for a model to make predictions on new unseen data.

Signup and view all the flashcards

Decision Trees

Classic ML models easy to understand.

Signup and view all the flashcards

Supervised Learning

Models learn from data and labeled outputs.

Signup and view all the flashcards

ML Data

A collection of numbers or text used to train machine learning models.

Signup and view all the flashcards

ML Model

A computer program that processes input and generates output, typically numerical.

Signup and view all the flashcards

ML Strengths and Weaknesses

How well a model learns is based on strengths and weaknesses.

Signup and view all the flashcards

Modelling Power

Ability of a model to discern complex patterns in data.

Signup and view all the flashcards

Data Requirements

Amount of data required to effectively train a model.

Signup and view all the flashcards

Trainability

Difficulty of training a model and expertise required

Signup and view all the flashcards

Interpretability

How easily we can understand a model and what it learns.

Signup and view all the flashcards

Linear Separability

Categories can be distinguished with one line.

Signup and view all the flashcards

Perceptrons (1969)

An influential book discussing decision making limitations.

Signup and view all the flashcards

Generative Learning

A model that generates new data that is similar to the training data.

Signup and view all the flashcards

Choosing a Model

Models are complex and have parameters with a cost function and hyperparameters.

Signup and view all the flashcards

Dall-E 3

Generative models that generate images from text prompts.

Signup and view all the flashcards

Context Window

A subset of a text source used by LLMs to predict the next word.

Signup and view all the flashcards

Generative Learning Benefit

Models that can learn from labeled and unlabeled data.

Signup and view all the flashcards

GANs (Generative Adversarial Networks)

Frameworks that combine generative and supervised learning.

Signup and view all the flashcards

Sequence generation

Model predicts the next thing that will happen in a sequence.

Signup and view all the flashcards

Study Notes

  • AI systems can encode biases if not taken care of.
  • Dall-E, an AI image generation model, produces images reflecting stereotypes of successful people as white, male, young, dressed in Western business attire, working in urban offices, and having common hairstyles.

Announcements

  • Assignment #2 is live and due Next Tuesday, with three possible submissions.
  • Three submissions are allowed, to mitigate accidental submissions, internet outages, etc.
  • There will be no feedback provided upon submitting.
  • There will be no feedback provided upon submitting.
  • Asignment #2 is due tonight
  • Asignment #1 grades have been released
  • Asignment #3 opens tonight
  • Asignment #3 is due the following Tuesday and will be like A2. It will be a multiple choice quiz in Python

Prerequisites and Learning Objectives

  • From last lecture, learn the building blocks of machine learning, data and models.
  • Data is the information used to train the model, like text and images
  • The model is a computer program that processes input and creates output that's usually a number or a collection of numbers.
  • Supervised learning learns to imitate, and learns to predict the label of data, which are created by human labelers
  • Key learning objectives for this content include:
  • Listing the strengths and weaknesses of different models.
  • Distinguishing between generative and supervised learning
  • Listing examples of generative learning systems and what they are used for

Machine Learning Building Blocks

  • Machine learning has 2 building blocks, data and models
  • Supervised learning is the basic paradigm
  • The goal is to list and describe the building blocks of ML systems
  • The goal is to identify problems that can be thought of as supervised learning

AI Caution, Fear, Excitement

  • “Success in creating effective AI, could be the biggest event in the history of our civilization. Or the worst. We just don't know. So we cannot know if we will be infinitely helped by AI, or ignored by it and side-lined, or conceivably destroyed by it,” ~Stephen Hawking
  • “The rise of AI will free people up to do things that software never will—teaching, caring for patients, and supporting the elderly, for example.” ~Bill Gates
  • “AI is far more dangerous than nukes.” ~Elon Musk
  • There are three reasons why people fear AI
  • Cynicism is the belief that it is rational not to cooperate
  • Humanism/racism is systematic bias against machines, denial of their potential moral worth and personhood
  • Conservatism is the fear of change, fear of the other tribe
  • None of these fears reflect well on those who hold them- Rich Sutton (UoA)

AI Versus ML

  • AI mimics intelligence, it includes systems with goals that make decisions.
  • ML is a subset of AI where knowledge is extracted from data to build predictive models.
  • Expert system MYCIN is given as an example of AI without ML.
  • A simulated path-finding robot is another example of AI, not ML.
  • Spam filters, YouTube recommendations, Google Lens, and ChatGPT are examples of both AI and ML.
  • ML is at the core of most modern AI systems.

Deciding which models to use

  • All basic models have different strengths and weaknesses
  • Modelling Power is whether the model can learn complex patterns
  • Data requirements is the amount of data needed to train the model
  • Trainability is whether it is difficult to train, and may require ML Expertise
  • Interpretability is understanding what the model learns

Data as a building block

  • In ML, data trains models to distill information and knowledge.
  • Data is a collection of discrete or continuous values conveying information, quantity, quality, facts, and statistics.

Course Map

  • The topics in this deck will include the Machine Learning Building Blocks 2 as well as Generative AI Systems, and choose a model.

Lecture Topics

  • This lecture will cover how to choose a model, and another important type of paradigm in MI: Generative Learning
  • Generative Learning will be broken down into Imitating Data, generative learning systems (Dall-E), Generative Adversarial Networks, and Sequence Generation Models (ChatGPT)
  • There will also be discussion of the power of scaling with data and computation over scaling with people.

Machine Learning Pipeline

  • The steps are training and prediction.
  • Training extracts info/knowledge from data to build the model.
  • Prediction uses the model to answer an input query.
  • Labels in ML examples provide the model with the question it should answer.
  • The label is the correct answer/output/prediction for a given input.

YouTube Video Recommendation example

  • For YouTube video recommendations, input data might include user ID, current viewing video, watch history, liked videos, and disliked videos.
  • The output is a list of top-K recommended videos

Google Lens example

  • The input is an image like a 2x2 Rubik's cube,
  • The output provides links to online stores selling similar items via object detection

MNIST example

  • MNIST, a commonly used ML dataset for research, consists of handwritten digits from 0 to 9,
  • It's used in real-world applications to read ZIP codes in postal services and check amounts in bank accounts.

Predicting text articles and movie reviews

  • Language tasks include classifying news articles by topic, classifying movie reviews as positive or negative.
  • There is also text generation based on input text.

Language Data

  • Before Large Language Models, ML focused on language, labeled "Natural Language Processing" (NLP).
  • Large Language Models are huge Neural Nets with Billions of connections & neurons trained mostly on text data.
  • They then output sentences

Bag of Words

  • The "Bag of Words" representation converts text to numbers for processing in ML models.
  • The goal is to provide calculations
  • Word order and importance can matter in language data.
  • "Near" can have many word associations

Count-Based Models and Vector Creation:

  • Count-based models convert words to numbers based on relationships, building a data table with rows and columns for each word.
  • We increment counts based on observed proximity, creating a vector representing word relationships.
  • This list of numbers creates a number vector.
  • Count-based models can show how word meanings shift over time.

Machine Learning Models

  • The model output is a number or a collection of numbers.
  • Classification models output a category, or a discrete value.
  • Regression models output some value that is a continuous value of the input.
  • Models answer questions based on input data

Classification Model

  • In classification, the output is a category, or a discrete label.
  • Examples include digit recognition (0-9), object detection (cube), and sentiment analysis (positive/negative).

Regression Model

  • Regression models work give output with a continuous value of the input, such as scoring digit writing, pricing, or rating a movie.

Model Output

  • The output of these models can be a variety of categories and values

Chat GPT

  • In ChatGPT data refers to the information used to train the model, like large collections of text from the internet, books, and other written sources

Machine Translation

  • Machine translation involves generating text output based on a given input.

Large Language models

  • LLMs convert large amounts of text into manageable word vectors and use context windows to predict the next word.

Attention

  • LLMs learn to focus on key words through a process, "attention", improving understanding.

Supervised Learning

  • Supervised learning, an ML framework, learns from labeled data where each input pairs with a label or correct answer.
  • The labels provide descriptive tags or values.
  • Supervised learning takes direction from the people selecting the answers

Categories

  • Labels in classification are categories, like categories for digits, objects, or sentiment.

Regression

  • Labels in regression are numerical values, like scores, prices, or ratings.
  • Supervised learning models have no agency and cannot be better than the data they are given.

Imitation

  • Supervised learning systems and expert systems are imitation machines.
  • ML Models are much easier to update.

Features

  • Features, like temperature, precipitation, day of the week, and clothing, are numbers or categorical data inputted into the model.
  • Models should generalize predictions on new data.

Generalization

  • Generalization makes predictions on new unseen data

Decision Trees

  • To get around this, decision trees, a classical machine learning model, can be implemented to find trends in data
  • In this model temperature is noted against the data.
  • They can be used for both classification and regression.

Decision Trees Pros and Cons

  • Decision trees are noted as having high explainability but struggle against higher, more complex data implementation and learning
  • It is important to remember how and when the systems work for this, and those of you in INT-D 161 will use powerful ML models

Other Learning

  • Supervised learning is the most basic paradigm in ML but there are other learning systems.
  • Other learning systems include Generative and Reinforcement Learning.

Lecture Topic Summaries

  • Data and models are crucial ML building blocks.
  • Data encoding depends on data type.
  • Supervised learning imitates human answers.
  • Decision trees are explainable classical models.

Sample Basic Models

  • Artificial Neural Network: The model is composed of layers of interconnected nodes, or neurons, that process and transmit information, it is comprised of an Input Layer, Hidden Layers, and an Output Layer
  • Decision Tree: A decision tree is a structured model that uses a series of binary decisions to classify or predict outcomes based on input features or attributes, it navigates based on conditions like temperature or weather, it then derives certain conditions
  • Perceptron: A perceptron is a single-layer neural network that performs binary classification by applying weights to input features and using a threshold to make a decision, it includes inputs, a Heavyside Step-function, and an output

Studying That Suits You

Use AI to generate personalized quizzes and flashcards to suit your learning preferences.

Quiz Team

Related Documents

More Like This

AI Bias and Fairness Quiz
7 questions

Quiz on Fairness of AI Solutions

CostEffectivePrudence7779 avatar
CostEffectivePrudence7779
Understanding AI Bias
5 questions

Understanding AI Bias

EnchantingJubilation avatar
EnchantingJubilation
Ethics of AI: Bias and Privacy
10 questions
Explainable AI: Bias, Trust, and Law
61 questions
Use Quizgecko on...
Browser
Browser