AI Associate
How many neural networks do VAEs rely on to generate new data based on sample data and neural radiance fields (NeRFs)?
2
General AI
AI that's good at everything
The tendency for people to attribute human motivation, emotions, characteristics or behavior to AI systems
Anthropomorphism
A computer program that mimics the way human brains process information.
Artificial Neural Network (ANN)
Computers are great at handling lots of data and doing complex calculations quickly. Humans are great at understanding context, finding connections between things even with incomplete data, and making decisions on instinct. What combines these two skillsets?
Augmented Intelligence
What type of AI Capability? Identify toxic comments on social media
Classification
What type of AI Capability? Identifying a plant from a picture
Classification
What is used to create a model?
Data
An advanced form of AI that helps computers become really good at recognizing complex patterns in data.
Deep learning
Refers to artificial neural networks being developed between data points in large databases.
Deep learning
What Einstein product gives agents intelligent, in-context suggestions, helping them do what they do best-help your customers?
Einstein Agent
What Einstein product has intelligent case routing, automatic triaging, and case field prediction?
Einstein Agent
What Einstein product can be used to increase deflection and reduce handle time?
Einstein Bots
What Einstein product helps managers take action with predictive service KPIs?
Einstein Discovery
What Einstein product can use pretrained models to classify text by the sentiment as either positive, neutral, or negative, and then be able to classify the underlying intent in a body of text?
Einstein Language
What Einstein product automates image classification?
Einstein Vision for Field Service
A framework that helps organizations assess and enhance their ethical practices in using AI technologies
Ethical AI Maturity Model
T or F: Generative AI uses statistical algorithms and machine learning to analyze data and make predictions.
False
T or F: Predictive AI is creative
False
Artificial intelligence that generates content, including text, images, video and music.
Generative AI
Field of artificial intelligence that focuses on creating new content based on existing data.
Generative AI
Technology that takes a set of data and uses it to create something new - like poetry, a physics explainer, an email to a client, an image, or new music - when prompted by a human.
Generative AI
Uses algorithms to analyze patterns in datasets to then mimic style or structure to replicate a wide array of content.
Generative AI
What is a huge part of NLP?
Generative AI
Like a detective. When it's shown pictures (or other data), it has to guess which are real and which are fake.
Generative Adversarial Network (GAN) discriminator
2 Types of Deep Learning Models
Generative Adversarial Network(GAN) Transformer
Made up of two neural networks: a generator and a discriminator.
Generative Adversarial Networks (GANs)
When two networks compete with each other, there is a generator creating an output based on some input and the discriminator trying to determine if the out put is real or fake. The generator then fine-tunes its output based on the discriminator's feedback, and the cycle continues until it stumps the discrimantor
Generative Adversarial Networks (GANs)
A neural network family that is trained to generate content.
Generative Pre-Trained Transformer (GPT)
What does GPT stand for?
Generative pre-trained transformer
An AI-based software tool that creates new content from a request or input.
Generator
Ensures that the system understands and relates to real-world knowledge, data, and experiences.
Grounding
When asked when is the best time to plant flowers, the AI response what type of flowers. What is this an example of?
Grounding
What is used to get a computer to notice trends in data so it can estimate where you assign inputs a weight and then make an estimate of the output based on those weights, then review the actual output, change the weights to make it more aligned to the correct results?
Guess-and-Check Method
This happens when generative AI analyzes the content we give it, but comes to an erroneous conclusion and produces new content that doesn't correspond to reality or its training data.
Hallucination
When generative AI gets a prediction wrong?
Hallucination
Common Concerns About Generative AI
Hallucinations Data Security Plagiarism User Spoofing Sustainability
The term used when we do not observe the layers between the inputs and the output
Hidden Layers
HITL
Human in the Loop
AI model architecture and extensive training data are two factors in the rapid improvement of generative AI, what is the 3rd?
Increase Parallel computing power
AI models are trained on massive amounts of data. What are these calleD?
LLM Large Language Models
What is LAVIS?
Language-Vision
What do you need more of for AI training to be considered deep learning?
Layers
What reduces words to their root?
Lemmatization
How computers can learn new things without being programmed to do them.
Machine Learning
When a computer learns from a limited or one-sided view of the world, and then starts making skewed decisions when faced with something new.
Machine Learning Bias
A program that's been trained to recognize patterns in data.
Model
What does the process of guess-and-check need to create in order to estimate the result?
Model
Element of natural language that re-elects how parts of words can be combined or uncombined to make new words.
Morphology
What type of AI Capability? ChatGPT
NLP
How has neural networks impacted NLP?
NLP has become faster and more contextually accurate.
Labels sequences of words and picks out the important things like names, dates, and times.
Named entity recognition (NER)
The ways human communicate
Natural Language
What is data processed the reverse way-from structured to unstructured?
Natural Language Generation
Field of artificial intelligence (AI) that combines computer science and linguistics to give computers the ability to understand, interpret, and generate human language in a way that's meaningful and useful to humans.
Natural Language Processing
What is it called when AI interprets everyday langage?
Natural Language Processing
What uses many techniques to interpret written or spoken language to understand the meaning and context behind it?
Natural Language Understanding
NLP's 2 subfields
Natural Language Understanding Natural Language Generation
Describes a machine's ability to understand what humans mean when they speak as they naturally would to another human.
Natural language processing (NLP)
Refers to systems that handle communication between people and machines.
Natural language understanding (NLU)
A web of connections, guided by weights and biases
Neural Networks
What are important tools for training AI models?
Neural Networks
What type of AI Capability? Predict your net quarter's sales
Numeric Prediction
Main Types of AI Capabilities
Numeric Predictions Classifications Robotic Navigation Language Processing
Breaking down text or speech into smaller parts to classify them for NLP
Parsing
If you ask generative AI what is its favorite color and it responds "blue", this is an example of what?
Prediction
Uses statistical algorithms to analyze data and make predictions about future events.
Predictive AI
Figuring out how to ask a question to get exactly the answer you need.
Prompt Engineering
One way to protect against hackers and harmful outputs is by being proactive about what terms and topics you don't want your machine learning model to address.
Prompt defense
Assigns a group to test a system or process for weaknesses.
Red-Teaming
Technique that teaches an AI model to find the best result via trial and error, as it receives rewards or corrections from an algorithm based on its output from a prompt.
Reinforcement Training
What type of AI Capability? AI-powered cars
Robotic Navigation
What takes larger texts and divides it into smaller, meaningful chunks?
Segmentation
Analysis used to determine whether a piece of text expresses a positive, negative, or neutral sentiment?
Sentiment
Determining the emotional tone behind words to gain an understanding of the attitudes, opinions, and emotions of a speaker or writer.
Sentiment Analysis
What Einstein product is preintegrated with Salesforce and your existing service channels, and comes out of the box with an intuitive user interface?
Service Cloud Einstein
What helps reduce the variations of word forms?
Stemming
Where words are reduced to their root form.
Stemming
Spreadsheet data is an example of what?
Structured Data
What type of data is used with supervised learning?
Structured Data
Generative AI
Subcategory of AI that takes words and turns them into unique images, sounds, and other words.
Tasks LLMs do
Summarization Translation Error Correction Question Answering Guided Image Generation Text-to-Speech
When a model learns from examples.
Supervised Learning
Where sentences are split into individual words, called tokens?
Tokenization
An umbrella term that describes a variety of offensive, unreasonable, disrespectful, unpleasant, harmful, abusive, or hateful language.
Toxicity
What type of Deep Learning Model is ChatGPT?
Transformer
Create outputs based on sequential data (like sentences or paragraphs) rather than individual data points.
Transformer Models
What model helps the model efficiently process context and is why it's used to generate or translate text?
Transformer model
T or F: Generative AI uses complex algorithms and deep learning to generate new content based on the data it is trained on.
True
What type of data is used for unsupervised learning?
Unstructured Data
News articles is data that is an example of what?
Unstructured data
The process of letting AI find hidden patterns in your data without any guidance.
Unsupervised Learning
What is VAE?
Variational AutoEncoders
What is used to create 2D and 3D images?
Variational AutoEncoders (VAEs)
Means that prompts and outputs are erased and never stored in an AI model.
Zero data retention
When models are trained on progressively more difficult tasks, so they will improve their ability to learn.
Zone of proximal development (ZPD)
The ________________'s job is to get better at telling real from fake, while the _________________ tries to get better at creating fakes.
discriminator, generator
The two networks compete with each other, with the ___________________ creating an output based on some input, and the ___________________ trying to determine if the output is real or fake.
generator, discriminator
The ___________________ then fine-tunes its output based on the _____________________ feedback, and the cycle continues until it stumps the ________________.
generator, discriminator's, discriminator
A type of deep learning model, and are especially useful for processing language. They're really good at understanding the context of words in a sentence because they create their outputs based on sequential data (like an ongoing conversation), not just individual data points (like a sentence without context).
transformer
