| … | … | @@ -11,3 +11,22 @@ The first attempt at structure learning is the github project |
| 11 | 11 | https://github.com/opencog/learn and perhaps this diary entry belongs |
| 12 | 12 | there. However, I want to incorporate the new ideas here, and so, for |
| 13 | 13 | now, this chapter is here. |
| 14 | |
| 15 | The language learning started with computing pair-wise MI between words, |
| 16 | then uusing this to create a network between words, then breaking up the |
| 17 | network into jigsaws. The jigsaws were then clustered, and the clusters |
| 18 | then used to parse the same text, again. This all worked fine. The next |
| 19 | step of the plan was to then repeat: find correlations between these |
| 20 | clustered jigsaws, then tile again, etc. ad ininfinitum. That step was |
| 21 | never taken, for two reasons: |
| 22 | |
| 23 | * Managing the batch processing pipeline became overwhelming. The |
| 24 | propsoed solution to that was to create sensori-motor agents that |
| 25 | could crawl the data "of their own volition"; or rather, automated |
| 26 | in such a way that I could aim agents at the data and have them |
| 27 | process and digest without my explicit intervention at a low level. |
| 28 | In particular: no more shell scripts, no more perl scripts; it would |
| 29 | all be Atomese. Solving that problem give rise to this github repo |
| 30 | that this design file is in right now. |
| 31 | |
| 32 | * The matri |