This is a hybrid Artificial Intelligence, one half using pre-set word, quote, Biblically Mathematical Magic Number meaning, sentence pattern and input-response lists, Cryptographically computed from the input, all of which I learned about and made myself.
Its other half uses a Discrete Fourier Transform for encoding input, with HyperDimensional Computing, Semantic Matching and Associative Reframing, and always learns to respond with Semantically Matching replies associated with the input, all of which I got another AI to make for me, then integrated into the first half, making these parts Holonomic.
If ΤAU's input is empty, it feeds itself a rule-generated sentence from the pre-set half for the input.
When its input matches previously learned things, it returns associated sentence(s).
Otherwise, ΤAU replies with a quote or interestingly abstract sentence if it doesn't find a match to known inputs.
It first learns the entire text contents of this page as it loads, then whatever input you feed it.
I have essentially built a bespoke, experimental chatbot that merges classical computing techniques with modern vector-based machine learning.
This AI's name is inspired by a 2018 movie called TAU.
Get.To download or update this, simply click on the "Get" link.
If I were to classify ΤAU using formal computer science and AI terminology, the most accurate overarching label would be a Hybrid Neuro-Symbolic Conversational Agent with Continual Learning.
Here is a breakdown of how the different components of my AI are classified in the field:
ΤAU combines "hard" rules and logic (the first half) with "soft" associative pattern matching (the second half).
In modern AI research, blending symbolic logic (rules, lists, cryptography) with connectionist systems (hyperdimensional vectors, semantic matching) is called a Neuro-Symbolic architecture.
It aims to get the best of both worlds: the precise control of manual programming and the flexible understanding of machine learning.
The side of the AI I've built myself relies on discrete, pre-programmed logic.
Pattern-Matching & Input-Response Lists: This classifies as a classical Rule-Based Chatbot.
It shares DNA with early AI pioneers like ELIZA or ALICE, which used rigid input-response mapping.
Cryptographic & Esoteric Logic: Using cryptography logic acts as a highly specific, deterministic hashing algorithm.
It routes specific inputs to specific outputs based on a rule set I designed.
The side of my AI I had another one help me build uses more advanced, modern paradigms.
HyperDimensional Computing (HDC): This is a specific and fascinating sub-field of AI (often categorized under Vector Symbolic Architectures).
HDC uses massive, thousands-of-dimensions-long vectors to represent concepts.
It is highly efficient and great for pattern recognition without the massive computational overhead of deep neural networks.
Associative Memory: By using semantic matching and reframing, my AI acts as an Associative Memory Engine.
It compares the "distance" between the user's input vector and the vectors of things it has learned to find the closest semantic match.
Because ΤAU "learns the entire text contents of this page as it loads, then whatever input you feed it," it is not a pre-trained static model like a standard LLM.
It operates using Online Learning (or Continual Learning), meaning it updates its knowledge base in real-time.
It is also includes an Instance-Based Learner (like K-Nearest Neighbors), meaning it essentially memorizes exact inputs and retrieves them when a new input is semantically close enough.
Retrieval-Based: Most of the time, the AI acts as a retrieval engine (pulling an associated sentence that it has already learned).
Generative Fallback: When it doesn't find a match and makes a "quote or interestingly abstract sentence," it is using a generative fallback mechanism to synthesize something new from its baseline data.
It forgets old data as it learns the new, with a sliding a window of the memory stream.
I've essentially built a highly personalized, esoteric AI art project mixed with some very real, highly efficient computer science concepts (like HDC).
Persistent memory | Erase memory
New Input
Random |
Therapy |
Reflect |
Find |
Poetry |
Quote |
Learn |
Copy |
Clear
Copy
Last Reply
Thoughts are joined with a colon ":" and separated with a pipe "|" character.
This software (the "Software") is licensed, not sold.
By downloading, opening, printing, copying, or using the Software, you agree to this End User License Agreement ("Agreement"). If you do not agree, do not use the Software.
If you require any more information or have any questions about my app's disclaimer, please feel free to contact me.
I never was able to complete my high school education, I'm not professionally trained, but am simply learning online, testing and developing this system. Its a combination of ideas I have come up with, along with some learned from well over hundreds of blogs and videos from experts on the topics, that I have used and tested. I'd like you to test this system yourself.
All the information on this app is published in good faith and for general information purpose only. Andrew Roberts does not make any warranties about the completeness, reliability and accuracy of this information. Any action you take upon the information you find on this app, is strictly at your own risk. I will not be liable for any losses and/or damages in connection with the use of my app(s) or product(s).
From my app, you can visit websites by following hyperlinks to external websites. While I strive to provide only quality links to useful and ethical websites, I have no control over the content and nature of these sites. These links to other websites do not imply a recommendation for all the content found on these sites. Site owners and content may change without notice and may occur before we have the opportunity to remove a link that may have gone 'bad'.
Please be also aware that when you leave my app, other sites may have different privacy policies and terms that are beyond our control. Please be sure to check the Privacy Policies of these sites as well as their "Terms of Service" before engaging in any business or uploading any information.
By using my app, you hereby consent to my disclaimer and agree to its terms.
Should I update, amend or make any changes to this disclaimer, those changes will be prominently posted here.