Glossary of AI spring related terms for the uninitiated laymen late 2024 version
Going to dabble in this learn in public trend/philosophy, albeit belatedly. Highly truncated panorama of the technologies, philosophies and players in the space.
Ok off the dome from memory;
Artificial intelligence; A combination or Machine leraning, Big data and Neural Networks to mimic human intelligence and solve problems
ArXiv; Most highly cited AI research papers and other research are posted here first.
Alignment; The iterative process of making the results or outputs of our communication with AI systems interpretable and predictable; minimizing the opaque black boxedness of these systems and reducing edge cases/ black swan events that can emerge from sufficiently trained rogue AI, etc.
Automation
Ai Agent
Big Data
Cloud Virtualized Compute, Storage, Memory, and Queuing/Message resources hosted globally and flexibly in state of the art safe Colocation Facilities by major companies around the world rather than consuming resources on your local device. Usually fully managed elastic and responsive to high throughput and demand.
Data Lake A massive (multiple Petabytes) data store of unstructured noSQL data
Dark Warehousing The inevitable endpoint of 3PL and warehousing using machine vision, Automated Guided Vehicles, and Deep learning from Warehouse Management systems based on trillions of interactions with IOT devices.
Compiler/Interpreter
ChatGpt Open AI
Context Windows
Diffusion/Latent Space
GAN Adversarial Networks are a cousin technology to LLMs without the Transformer element. Basically Gan is a type of neural network that weights an array of pixels and through bombaredment of several layers of network mesh and “extract” the shapes associated with different metadata
GPT
Glaze/Nightshade; two different technologies that add noise to images
Google Deepmind
Grok; Your familiar with this model if you use X
Hallucinations Anything inconvenient that an LLM does.
huggingface.co; LLM model leaderboard with different sub metrics
(O)Llama Open source LLM you can host this software on your own device
LLM Large Language Model, large natural language models that predict the most logical follow up tokens using context from the internet and tokens in the prompt. although some tests demonstrate that sufficiently sized models have emergent reasoning and not just regurgitative language prediction ability
LangChain: A centralizing process management service that feeds your personal or enterprise data, storage, files and appliances compiling them into containers/vectors and plugging them into different LLMS and extruding the results with AI agents.
Lobotomized AI; Various Ai can be temporarily intentianally “lobotomized” or made less effective to save resources or for political reasonsV
Longevity Escape Velocity/Protein folding problem. Probably the single biggest driver for the accelerationists in the early days
Machine Learning
Metadata: tags and annotations that you apply to files for searchability and readability by machines
Moloch
MoroveC’s Paradox; The once measurable but increasingly mostly perceived lack of innovations in the material sciences, cybernetic and robotics fields vis a vis the software/AI startup environment. Or otherwise an inverse relationship between technology being agentic and evolving navigating a physical world AFTER developing a “brain” whereas humans evolutionary trajectory was the opposite.
Paperclip Maximizer The meme that an improperly aligned AI connected to the wrong device with the wrong permissions could misinterpret a command and turn all matter into paperclips
P(doom): Chance of worst possible scenarios for humankind from AI. The singular worst option.
Roko’s Baskalisk; A SciFi/reallife scenario that you probably dont want to research. This isn’t reverse psychology. You might be literally hexxed! Bad juju.
Singularity/SuperIntelligence; The point in which an AI or android; embued with sapience/consciousness becomes self driving and is more intelligent than all humans on the planet combined.
Sovereign AI
Tokens
Training Data
Techno-feudalism; Yanis Varoufakis Book detailing the relationship between consumers and databrokers that extract data as a form of economic rent
UBI;
Suno/Udio: Generative AI for song generation. Having tried the free version of both I prefer Suno personally, but milage may vary.
Useless eaterism; coined by Homo Deus author Yuval Noah Harari describing a state of dereliction the majority of the unskilled masses will find themselves in.
Quines; Lex Freidman made a great video about this concept; basically how to make lines of code recursively reproduce themselves.
Ok and out of curiosity and for fun this is a list of terms GPT 4o gave me. pretty distinct responses. Prompt; “Can you give me a glossary of ai related terms?”
AI Terms by AI
Artificial Intelligence (AI): The simulation of human intelligence processes by machines, particularly computer systems. These processes include learning, reasoning, and self-correction.
Machine Learning (ML): A subset of AI that involves training algorithms to recognize patterns and make decisions based on data without being explicitly programmed.
Deep Learning: A class of machine learning algorithms based on neural networks with many layers (deep neural networks), often used for tasks like image and speech recognition.
Natural Language Processing (NLP): A field of AI focused on the interaction between computers and humans through natural language, including tasks like translation, sentiment analysis, and text generation.
Neural Network: A series of algorithms that attempt to recognize underlying relationships in a set of data through a process that mimics the way the human brain operates.
Supervised Learning: A type of machine learning where the model is trained on labeled data, meaning the outcomes are already known.
Unsupervised Learning: A type of machine learning where the model is trained on unlabeled data and must find patterns and relationships on its own.
Reinforcement Learning: A type of machine learning where an agent learns to make decisions by performing actions and receiving rewards or penalties.
Algorithm: A set of rules or processes for solving a problem or performing a task, often used in AI for processing data and making decisions.
Training Data: The dataset used to train a machine learning model, which helps the model learn the patterns and relationships in the data.
Spring Framework Terms
Spring Framework: A comprehensive framework for building Java-based enterprise applications, offering support for various aspects of application development including dependency injection, transaction management, and web applications.
Dependency Injection (DI): A design pattern used in Spring to manage object dependencies, allowing objects to be injected into other objects rather than being created by the objects themselves.
Inversion of Control (IoC): A principle where the control of object creation and lifecycle is transferred from the application code to the Spring container.
Spring Boot: An extension of the Spring Framework that simplifies the setup and development of new Spring applications by providing pre-configured templates and auto-configuration.
Spring MVC: A module of the Spring Framework that provides a Model-View-Controller architecture and is used to build web applications.
Bean: An object managed by the Spring container. Beans are created, managed, and destroyed by the Spring IoC container.
Application Context: The central interface to the Spring IoC container, responsible for managing the lifecycle of beans and providing configuration information.
Aspect-Oriented Programming (AOP): A programming paradigm supported by Spring that allows for separating cross-cutting concerns (such as logging and transaction management) from the main business logic.
Spring Data: A set of libraries and projects within the Spring ecosystem aimed at simplifying data access and integration with various data sources (e.g., databases, NoSQL systems).
Spring Security: A module of the Spring Framework that provides authentication, authorization, and protection against common security vulnerabilities.
Spring Cloud: A set of tools and frameworks for building and deploying microservices-based applications in a cloud environment.
Spring Batch: A framework for building robust batch processing applications, including tasks like reading, processing, and writing large volumes of data.
This was a fun exercise. Its a glossary so i’m not obliged to give definitions for every entry. Which do you like better?