GenAI - Generate thoughts

GenAI - Generate thoughts

Let me start with something very clear - This article is not generated by GenAI :) It is created by Kishore Chivukula to share some of the foundations, learnings, resources and interest in GenAI space.

The purpose as the title says is to help you as a reader, "Generate thoughts" with respect to GenAI technology, tools, services and buzz in simple and informative manner providing clarity, references to in-depth articles and research material.

It would take a good 2 or 3 hours read and don't forget to keep your coffee ready :)

What is GenAI?

Generative AI (GenAI) is a type of Artificial Intelligence that can create a wide variety of data, such as images, videos, audio, text, and 3D models.

Want to know more, check this link - https://generativeai.net/

What is ChatGPT?

ChatGPT, which stands for Chat Generative Pre-trained Transformer, is a large language model-based chatbot developed by OpenAI, notable for enabling users to refine and steer a conversation towards a desired length, format, style, level of detail, and language used.

Even though many use "ChatGPT" word interchangeably to "GenAI" you now know that ChatGPT is one of GenAI applications and may be the name made it very popular that it is used everywhere.

What is the backbone for GenAI?

Foundation models - Foundation models are AI models designed to produce a wide and general variety of outputs. They are capable of a range of possible tasks and applications, such as text, image or audio generation.

Data - Vast amounts of data is used to train these models via crawling and proprietary data wherever applicable

Data is available in Web and search engines already indexed them, then how are LLMs different?

Historically, AI models had been focused on perception and understanding.?

However, large language models, which are trained on internet-scale datasets with hundreds of billions of parameters, have now unlocked an AI model’s ability to generate human-like content. Models can read, write, code, draw, and create in a credible fashion and augment human creativity.?

Follow-up is, How Do Large Language Models Work?

Large language models are trained using unsupervised learning. With unsupervised learning, models can find previously unknown patterns in data using unlabelled datasets. This also eliminates the need for extensive data labeling, which is one of the biggest challenges in building AI models.

Check this informative article on LLMs - https://www.nvidia.com/en-us/glossary/data-science/large-language-models/

What are parameters?

The given parameters of a particular model represent the weight of the various probabilities that it can produce. They are technically defined as variables whose values are adjusted during training to establish how input data gets transformed into the desired output. A single parameter is thereby a value that is earned and adjusted by an artificial intelligence algorithm during the training process to make decisions and predictions.

https://www.profolus.com/topics/what-are-parameters-in-ai-models/

We hear terms "Foundational Models" and also "Large Language Models", are they same?

A large language model (LLM) is any statistical model of language built on “large” swaths of data and “a lot” of parameters. “Large” and “a lot” is compared to previous approaches just a few years ago; think terabytes of training data rather than megabytes.

A foundation model is a term introduced by Stanford and the Stanford Center for Research on Foundation Models (CRFM). Although the paper isn’t entirely clear, characteristics of a foundation model as introduced in this paper include:

  • Large Language Modeling is used. However, it can be for a model with just language abilities or for models handling language-images or language-something else. The commonality across these models is the use of modern LLM approaches, among other things.
  • They are trained using self-supervision.
  • They can function as a “pretrained” model for further fine-tuning.
  • They are trained on unconsented data that may be in violation of copyright/license.

What are some of those examples for Foundation models?

Check this insightful article - https://www.adalovelaceinstitute.org/resource/foundation-models-explainer/

How to use GenAI tools for my personal needs?

Check below link and it covers useful tools like ChatGPT and many other options which augments your creativity and productivity.

They help you to acquire new skills quickly like creating marketing videos, mobile apps with very less learning curve.

https://scribehow.com/library/generative-ai-tools

How to go about developing business applications and customer centric solutions?

What is Fine-Tuning in Generative AI?

Fine-tuning is a technique in which pre-trained models are customized to perform specific tasks or behaviors.

It involves taking an existing model that has already been trained and adapting it to a narrower subject or a more focused goal. Using this technique, you can use your OWN datasets and get customized results. This customization step lets you get more out of the service by providing:

  • Higher quality results than what you can get just from prompt design
  • The ability to train on more examples that can fit into a prompt
  • Lower-latency requests

Check this link - https://innodata.com/quick-concepts-fine-tuning-in-generative-ai/

What are GenAI use-cases that industry has identified and investing?

They fall under three categories

  • Quickly automating and simplifying project workflows.
  • Taking repetitive tasks off the plates of busy employees.
  • Helping businesses maintain high-quality and high-volume production standards.

https://www.eweek.com/artificial-intelligence/generative-ai-enterprise-use-cases/

Technology makes me feel I can do anything using GenAI services, do I have to know anything specific?

Yes, 6 important things and they would help you a lot in making right choices

  1. Problem statement - What problem are you solving for your customer or end-user
  2. Cost - This could be super expensive and you should understand for your use-case and usage how much does it cost
  3. User experience/validation - Your end-user has to interact with your chatbot, voice assist or summarization feature, so make sure you embed it into their workflows and make it easy, intuitive
  4. Security and Privacy
  5. Intellectual property (IP) rights
  6. Transparency and explainability

Insights from OpenAI

OpenAI in below blog shares tons of things that they learnt across 22 countries and how they are planning to take certain things forward

https://openai.com/blog/insights-from-global-conversations

Below links are super useful to build general awareness on how GenAI is being envisioned across industries, it's caveats and some of expert findings or recommendations

Closing comments

I am closing :) Hope you find this article useful.

Feel free to share it to others.

要查看或添加评论,请登录

Kishore Chivukula的更多文章

  • POC - Proof of Clarity

    POC - Proof of Clarity

    The fundamental objective of any POC, Proof of Concept, is to provide a business, group or team ABSOLUTE CLARITY on how…

  • My Important Experience as a Developer..

    My Important Experience as a Developer..

    There is lot of talk these days on Developer Experience, Developer Productivity and Developer Satisfaction. Many top…

  • Are you an engineer with these skills?

    Are you an engineer with these skills?

    All engineers software, factory, mechanical, industry, electrical, communication, network construction, quality…

  • 17 years - 17 Learnings

    17 years - 17 Learnings

    I completed 17 years this week in software industry and even though everyone says it, I am going to still say "It still…

  • GIDS - Day 3 Learnings

    GIDS - Day 3 Learnings

    I was lucky to attend Day 3 of Great International Developer Summit that happened in "Namma Bengaluru" from 23-26 April…

    1 条评论
  • AWS re:invent 2023 - Learnings

    AWS re:invent 2023 - Learnings

    I have this good habit of following re:invent from AWS virtually (may be one day it would be in person!) for the last 5…

  • Open Source + GenAI learnings

    Open Source + GenAI learnings

    This article covers learnings, insights and pointers that I heard in "Open Source India" 20th edition conference in…

  • Cloud is changing. Are you?

    Cloud is changing. Are you?

    If you are someone passionate about cloud ecosystem and carefully following news, updates in the last 6-9 months from…

    2 条评论
  • Journey of Cloud 6

    Journey of Cloud 6

    You would have heard of #CLOUD9 but this article talks about #CLOUD6 :) I will try my best to ensure it's worth every…

  • One pager - AWS re:invent 2022 Summary

    One pager - AWS re:invent 2022 Summary

    First things first. Missed any keynote? Watch them here :) Regarding lots of announcements (BIG, MEDIUM, SMALL across…

社区洞察

其他会员也浏览了