I sent a message to Anton saying "have you tried skipping the MST step entirely?" and "does the quality of your results degrade if you do?" I kind of expect the reply to be "huh, what do you mean by that?" but sure, that's off-topic for this list. --linas
p.s. re computational biology; do not underestimate the complexity of biology. People have actually built multi-million-$$ protein-folding hardware, and protein-folding@home is still a thing. You may be able to use machine-learning techniques to tease out metabolic pathways, discern the structure in them (I currently believe they're not unlike natural language in structure) but there's a terrible combinatorial explosion in biochem that should not be taken lightly. --linas On Sat, Feb 9, 2019 at 8:29 PM Ben Goertzel <b...@goertzel.org> wrote: > Current MST step is suboptimal, but there is value in extracting a > parse tree with no-links-crossing from the tangly graph of dependency > links interconnecting the words in a sentence.... As you now this > sort of pruning is key to how language works ... syntax "rules" exist > to enable this sort of pruning, thus allowing us to pinpoint exact > sets of semantic relationships within the broader tangly web of > semantic relationships that our mind identifies between any set of > concepts or objects or relationships... > > However, this is an in-depth technical discussion probably best not > carried out on this general-purpose list... > > On Sun, Feb 10, 2019 at 3:58 AM Linas Vepstas <linasveps...@gmail.com> > wrote: > > > > > > > > On Sat, Feb 9, 2019 at 4:22 AM Ben Goertzel <b...@goertzel.org> wrote: > >> > >> > >> We are now playing with hybridizing these symbolic-ish grammar > >> induction methods with neural net language models, basically using the > >> predictive models produced by models in the BERT lineage (but more > >> sophisticated than vanilla BERT) in place of simple mutual information > >> values to produce more broadly-context-sensitive parse choices in > >> Linas's MST parser... > > > > > > This last sentence suggests that the near-total confusion about MST > continues to persist in the team. I keep telling them to collect the > statistics, and then discard the MST parse **immediately**. Trying to > "improve" MST is a total waste of time. > > > > Seriously: Instead, try skipping the MST step entirely. Just do not > even do it, AT ALL. Rip it out. It is NOT a step that the algorithm even > needs. I'll bet you that if you skip the MST step completely, the quality > of your results will be more-or-less unchanged. The results might even get > better! > > > > If your results don't change, by skipping MST, or if your results get > better, by skipping MST, then that should be a clear indicator that trying > to "improve" MST is a waste of time! > > > > -- Linas > > > > -- > > cassette tapes - analog TV - film cameras - you > > Artificial General Intelligence List / AGI / see discussions + > participants + delivery options Permalink > > -- > Ben Goertzel, PhD > http://goertzel.org > > "The dewdrop world / Is the dewdrop world / And yet, and yet …" -- > Kobayashi Issa -- cassette tapes - analog TV - film cameras - you ------------------------------------------ Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/Ta6fce6a7b640886a-M9a77ba22548daeb0cb7f9c62 Delivery options: https://agi.topicbox.com/groups/agi/subscription
