CS362 - Final Flashcards by Sultan AlFadhli

Q

Word Embedding

A

Representing words as continuous-valued vectors in lower dimensions.

Week 6: Deep Learning for Natural Language Processing

How well did you know this?

1

Not at all

2

3

4

5

Perfectly

Q

Sequence-to-Sequence (Seq2Seq)

A

Models that map input sequences to output sequences.

Week 6: Deep Learning for Natural Language Processing

How well did you know this?

1

Not at all

2

3

4

5

Perfectly

Q

Transformer

A

A deep learning model using self-attention for sequence tasks.

Week 6: Deep Learning for Natural Language Processing

How well did you know this?

1

Not at all

2

3

4

5

Perfectly

Q

Text Preprocessing

A

Steps like tokenization, lowercasing, stemming for preparing text.

Week 6: Deep Learning for Natural Language Processing

How well did you know this?

1

Not at all

2

3

4

5

Perfectly

Q

Bag-of-Words

A

Text representation ignoring word order, using word frequency.

Week 6: Deep Learning for Natural Language Processing

How well did you know this?

1

Not at all

2

3

4

5

Perfectly

Q

Agent

A

An entity that perceives its environment and acts to achieve goals.

Week 8: Intelligent Agents & Reinforcement Learning

How well did you know this?

1

Not at all

2

3

4

5

Perfectly

Q

Environment

A

The external system an agent interacts with.

Week 8: Intelligent Agents & Reinforcement Learning

How well did you know this?

1

Not at all

2

3

4

5

Perfectly

Q

State

A

Current situation of the environment.

Week 8: Intelligent Agents & Reinforcement Learning

How well did you know this?

1

Not at all

2

3

4

5

Perfectly

Q

Action

A

Steps taken by the agent to transition states.

Week 8: Intelligent Agents & Reinforcement Learning

How well did you know this?

1

Not at all

2

3

4

5

Perfectly

Q

Policy

A

Mapping from states to actions.

Week 8: Intelligent Agents & Reinforcement Learning

How well did you know this?

1

Not at all

2

3

4

5

Perfectly

Q

Reward

A

Feedback received after an action.

Week 8: Intelligent Agents & Reinforcement Learning

How well did you know this?

1

Not at all

2

3

4

5

Perfectly

Q

Episode

A

A complete sequence of states, actions, and rewards.

Week 8: Intelligent Agents & Reinforcement Learning

How well did you know this?

1

Not at all

2

3

4

5

Perfectly

Q

Passive RL

A

Learning utility of states with a fixed policy.

Week 8: Intelligent Agents & Reinforcement Learning

How well did you know this?

1

Not at all

2

3

4

5

Perfectly

Q

Active RL

A

Learning optimal policy through exploration.

Week 8: Intelligent Agents & Reinforcement Learning

How well did you know this?

1

Not at all

2

3

4

5

Perfectly

Q

Generalization in RL

A

Ability of learned policies to work in new environments.

Week 8: Intelligent Agents & Reinforcement Learning

How well did you know this?

1

Not at all

2

3

4

5

Perfectly

Q

Robot

A

A physical agent that performs tasks in the physical world.

Week 9: Robotics

How well did you know this?

1

Not at all

2

3

4

5

Perfectly

Q

Effector

Study These Flashcards

A

Part of a robot (e.g., gripper) that interacts with the environment.

Week 9: Robotics

Q

Actuator

Study These Flashcards

A

Drives motion in robot components.

Week 9: Robotics

Q

Perception

Study These Flashcards

A

The robot’s ability to understand surroundings using sensors.

Week 9: Robotics

Q

Collaborative Robot (Cobot)

Study These Flashcards

A

Designed to work safely alongside humans.

Week 9: Robotics

Q

Social Robot

Study These Flashcards

A

Interacts with humans using social norms.

Week 9: Robotics

Q

Python

Study These Flashcards

A

Most common programming language in AI.

Week 10: AI Languages, Libraries, Tools, and Hardware

Q

NumPy

Study These Flashcards

A

Library for numerical computation in Python.

Week 10: AI Languages, Libraries, Tools, and Hardware

Q

PyTorch

Study These Flashcards

A

Deep learning framework by Meta.

Week 10: AI Languages, Libraries, Tools, and Hardware

TensorFlow

Google’s open-source deep learning framework. Week 10: AI Languages, Libraries, Tools, and Hardware

Scikit-learn

Library for ML algorithms (e.g., classification, clustering). Week 10: AI Languages, Libraries, Tools, and Hardware

XGBoost

Library for gradient boosting. Week 10: AI Languages, Libraries, Tools, and Hardware

AI Accelerator

Hardware designed to speed up AI tasks (e.g., neural networks). Week 10: AI Languages, Libraries, Tools, and Hardware

GPU

Graphics card used to accelerate deep learning computations. Week 10: AI Languages, Libraries, Tools, and Hardware

TPU (Tensor Processing Unit)

AI-specific chip developed by Google. Week 10: AI Languages, Libraries, Tools, and Hardware

MLOps

Practices to manage and automate machine learning development and deployment. Week 11: AI Development Life Cycle

Design-Develop-Deploy Phases

Stages of creating AI systems. Week 11: AI Development Life Cycle

Problem Framing

Understanding and defining the ML task. Week 11: AI Development Life Cycle

Model Training Workflow

Steps to train, validate, and optimize ML models. Week 11: AI Development Life Cycle

Deployment

Putting a trained model into a production environment. Week 11: AI Development Life Cycle

AI Ethics

Moral principles guiding the development and use of AI. Week 12: Ethics, Safety, and Future of AI

AI Safety

Ensuring AI systems operate reliably and without harm. Week 12: Ethics, Safety, and Future of AI

Weak AI

Limited-scope AI. Week 12: Ethics, Safety, and Future of AI

Strong AI / AGI

Human-level or beyond intelligence in machines. Week 12: Ethics, Safety, and Future of AI

AutoML

Automating the ML pipeline to make it accessible and efficient. Week 12: Ethics, Safety, and Future of AI

Future of AI

Involves applications in healthcare, finance, education, and more. Week 12: Ethics, Safety, and Future of AI