View source for AI & Machine Learning

=Chatbots and Conversational AI=

: Most famous at the time of this writing is [https://chat.openai.com '''ChatGPT'''], but it is not the first or only AI capable of holding a conversation. Months before it became publicly available, I had subscribed to [https://replika.com '''Replika'''], which offers a more limited experience but was still amazing in its ability to mimic a human. Now, Google has [https://blog.google/technology/ai/bard-google-ai-search-updates/ announced its own version, called '''Bard'''], but it is currently only available to certain Google developers. And a Google spin-off company called [https://deepmind.com ''DeepMind''] is working on another tool called [https://www.ai.nl/artificial-intelligence/deepminds-conversational-ai-chatbot-sparrow-shows-dialogue-difficulties-and-need-for-responsible-ai/ '''Sparrow'''], also expected to have a public release sometime in 2023.

=AI-powered Search=

: Microsoft's investment in [https://openai.com OpenAI], creators of [https://chat.openai.com '''ChatGPT'''] and its roots, GPT-3 and other language models, has led to [https://www.androidcentral.com/apps-software/microsoft-bing-edge-openai-chatgpt announcements regarding newly-beefed up applications] from that company. [https://bing.com '''Bing'''], the Microsoft mapping app, is now enhanced with chat AI capabilities, as is [https://www.microsoft.com/en-us/edge '''Edge'''], the browser currently offered by the company. According to their FAQ, one can access the new Bing at https://bing.com which now does answer short queries, with support for longer, more complex ones coming soon.  [https://google.com '''Google'''] has also said their search includes AI support to provide better answers to complex queries.

=AI for Education=

: [https://chat.openai.com '''ChatGPT'''] claims it can help one learn many things, but so far all I've gotten from it have been links to sites that offer language tutorials, often for a price. But in fairness, that's all I've asked for so far. OpenAI's website has been overloaded lately (following the Microsoft AI-enhanced '''Bing''' and '''Edge''' rollouts) so it has not been possible to see what else is available.

==AI Language Tutors==

: [https://spectrum.ieee.org/duolingo '''DuoLingo'''] is an AI-based language-learning app that uses a number of advanced techniques to provide a flexible and ever-improving experience that both engages and educates the learner simultaneously while also evaluating learner progress and adjusting the app's subject matter presentation.

=The Competition=

:* [https://venturebeat.com/ai/who-will-compete-with-chatgpt-meet-the-contenders-the-ai-beat/ Who will compete?]
:* [https://github.com/openai/openai-cookbook OpenAI's Cookbook]
:* [https://towardsdatascience.com/gpt-3-a-complete-overview-190232eb25fd A complete overview of GPT-3]
:* [https://www.anthropic.com/ Antrhopic AI]

=Definitions=

''Also see the [[Glossary]] on this site''

==Artificial Intelligence (AI)==
<q>the theory and development of computer systems that are able to perform tasks that normally require human intelligence, such as visual perception, speech recognition, learning, decision-making, and natural language processing</q>
:: -- [https://globalpolicy.ieee.org/wp-content/uploads/2019/06/IEEE18029.pdf IEEE.org]

===Open Source AI===

:* [https://openai.com OpenAI] an AI research and deployment company
:* [https://www.datamation.com/open-source/open-source-artificial-intelligence-50-top-projects/ 50 top open source AI projects]
:* [https://techcrunch.com/2022/12/02/openais-chatgpt-shows-why-implementation-is-key-with-generative-ai/ TechCrunch article about OpenAI's ChatGPT]
:* [https://www.datamation.com/artificial-intelligence/open-source-artificial-intelligence-leading-projects/ Leading open AI projects]

==Machine Learning (ML)==
<q>In general, ML is the use of computer algorithms that can learn and adapt from processing digital data.</q>
:: --Rohit Sharma in [https://www.amazon.com/dp/B0B662D7ZW Introduction to TinyML]

==Deep Learning (DL)==
<q>[a subfield of ML] that relies heavily on neural networks to process data and produce results</q>
:: --Rohit Sharma in [https://www.amazon.com/dp/B0B662D7ZW Introduction to TinyML]

==TinyML==
<q>Tiny machine learning is broadly defined as a fast growing field of machine learning technologies and applications including hardware, algorithms and software capable of performing on-device sensor data analytics at extremely low power, typically in the mW range and below, and hence enabling a variety of always-on use-cases and targeting battery operated devices.</q>
:: --[https://tinyml.org The TinyML Foundation]

<!--

-->

=References=
* [https://tinymlbook.com/ TinyML book] homepage (with videos & large sample PDF)
** [https://www.amazon.com/TinyML-Learning-TensorFlow-Ultra-Low-Power-Microcontrollers/dp/1492052043 TinyML] book on Amazon
* [https://spectrum.ieee.org/number-representation New thoughts on number representation for ML] - article
* [https://www.amazon.com/TinyML-Cookbook-artificial-intelligence-ultra-low-power/dp/180181497X TinyML Cookbook]
* [https://www.amazon.com/dp/B0B662D7ZW Introduction to TinyML]

<!--

-->