6 Matching Annotations
  1. Last 7 days
    1. we are using set theory so a certain piece of reference text is part of my collection or it's not if it's part of my collection somewhere in my fingerprint is a corresponding dot for it yeah so there is a very clear direct link from the root data to the actual representation and the position that dot has versus all the other dots so the the topology of that space geometry if you want of that patterns that you get that contains the knowledge of the world which i'm using the language of yeah so that basically and that is super easy to compute for um for for a computer i don't even need a gpu

      for - comparison - cortical io / semantic folding vs standard AI - no GPU required

    2. we basically grow models of let's say same quality like all the others by using thousand time or ten thousand times less training data

      for - comparison - semantic folding vs normal machine learning - training dataset sizes and times

    3. for - semantic folding - semantic fingerprint - cortical.io - numenta - sparse coding - Francisco Webber - symmathesetic fingerprint

      summary - In this informative interview, Francisco Webber, a principal cofounder of Cortical.io, discusses how the company's core technology, semantic folding and semantic fingerprints of words is unique and differs from the usual AI large language models. - Cortical I/O's approach is a biomimicry approach that is based on representing words in the way that brain operates. - It employs a word-to-geometry mapping implemented using Numenta's sparse coding technique. - This approach allows Cortical to train using very small training data sets of 100 gigabytes of data, which takes a few hours to train - many orders of magnitudes smaller than normal AI training data sets.

    4. if you have bitmaps let's say 100 times 100 in in square and you now throw in let's say 200 dots in this bitmap the rest is white you should what you need is a function that renders any given word in a bitmap such that words that are similar render in two similar bitmaps

      for - example - semantic fingerprint bitmap - adjacency - semantic fingerprint bitmap - semantic folding - symmathesetic fingerprint - symmathesetic folding - Indyweb - adjacency - indranet - salience mismatch

      example - semantic fingerprint bitmap - 100 x 100 square - 200 dots in the bitmap - sparse coding - function that renders words in the bitmap such that - words that are similar render in two similar bitmaps

      adjacency - between - semantic fingerprint - semantic folding - symmathesetic fingerprint - symmathesetic folding - Indyweb - Indranet - adjacency - salience mismatch - adjacency relationship - This word-to-geometry mapping is the key idea and can also be employed within Indyweb to represent the concept of word/idea adjacency unique to the meaningverse of each language user - While Cortical develops dictionaries for specific domains, within Indyweb, we can go even more granular, and develop dictionaries for each indyvidual!

      definition - indyvidual dictionary - In Indyweb, an indyvidual's dicitionary can be calculated by employing a word meaning-to-geometry bitmap to determine the adjacencies salient to any word - This can be used to reduce salience mismatch (misunderstanding) that is inherent in any human symbolic communication

    5. what we basically do is that we try to find a representation for textual content so we call these representation fingerprints and they are like bitmaps

      for - definition - semantic folding

      definition - semantic folding - geometric (bitmap) representation of textual content

  2. Feb 2024