teixi

18+ RossGayler , 3 months ago to cogsci

@cogsci
The notion of role-filler binding as central to cognitive(ish) representations has been around for ages (possibly under different names, such as slot-value in GOFAI: https://en.wikipedia.org/wiki/Frame_(artificial_intelligence)). This is hardly surprising because it's effectively the same as variable-value.

The role is generally treated as though it's an atomic symbol, whereas it's not uncommon for the filler to be taken as a composite value (e.g. a tree). I am toying with embracing the idea of roles also being composite representations.

In a cognitive-agent/robotic context, I think it might be useful for the role to be a "sensorimotor program" and the filler to be the sensory input arising from running the sensorimotor program specified by the role. (This is heading towards a Predictive State Representation: https://en.wikipedia.org/wiki/Predictive_state_representation).

(1) I would greatly appreciate any pointers to discussions of role-filler bindings as sensorimotor predictions (similar or related to the sense above).

"Attention" could be construed as a "run/don't_run" flag in the sensorimotor program. This is basically treating attention as a kind of action and "don't attend" as not doing that action. (If that were true it's possible that there may also be other attention mechanisms, e.g. the precision weighting posited by Predictive Coding: https://en.wikipedia.org/wiki/Predictive_coding#Precision_weighting).

(2) I would greatly appreciate any pointers to discussions of attention as a kind of executable sensorimotor action.

#CogSci #CognitiveScience #CognitiveRobotics #CognitiveNeuroscience #neuroscience

18+ teixi , 3 months ago

@RossGayler @cogsci

Not so fast ;)

ie check:

Structure Learning in Predictive Processing Needs Revision
@Iris ea
https://doi.org/10.1007/s42113-022-00131-8

Downstream network transformations dissociate neural activity from causal functional contributions
@kordinglab ea
https://doi.org/10.1038/s41598-024-52423-7

18+ teixi , 3 months ago

@RossGayler @cogsci @Iris @kordinglab

speak same lang, diff dialect ;)

1 GOFAI identifier/filler
DMN Decision Model Notation https://www.omg.org/spec/DMN
PMML Predictive Model Markup Language
https://dmg.org/pmml/
Drools/KiE
https://kogito.kie.org/guides/
https://learn-dmn-in-15-minutes.com/

2 Intelligence without Representation:
A Historical Perspective
https://doi.org/10.3390/systems8030031
https://typeset.io/papers/intelligence-without-representation-w8cznii0l5

HTH

18+ teixi , 3 months ago

@RossGayler @cogsci @Iris @kordinglab

upps... you're right I went for GOFAI role/filler but without relation to sensimotor contexts...

Interesting R Sutton —the bitter lesson— now this: They don't have a life! Expert Systems ;) yet the Summary slide, IMHO non-conclusive & confusing

Now searching further into it —still curious— perhaps these:

1/2

18+ teixi , 3 months ago

@RossGayler @cogsci @Iris @kordinglab

Situated Embodied Artificial Intelligence SEAI
https://doi.org/10.1080/09540090701192584

Sensorimotor Contingencies Theories SMCT
https://www.diva-portal.org/smash/get/diva2:1208897/FULLTEXT01.pdf
https://www.zora.uzh.ch/id/eprint/114184/1/PhD-Thesis_Published_NicoSchmidt_el.pdf
https://doi.org/10.3389/fnbot.2022.847054

2/2

HTH

aheadofthenerve , 6 months ago to random

I guess we do #introduction posts over here? I work on the #neuroscience (am I doing those hashtags right!?) of learning and memory, specifically how we learn while we navigate space and context. To do this, I take in vivo recordings (currently calcium imaging but ephys has my heart) of freely moving rats! After that, I use computational and mathematical approaches to analyze their neural activity! I am currently a BRAIN Initiative K99/R00 postdoc at Northwestern working with John Disterhoft and Sara Solla. I was trained at MIT with Matt Wilson, where I got my PhD in biology, and my BS is from Carnegie Mellon. Welcome!

teixi , 5 months ago

@aheadofthenerve

You can find #neurofellows & add yourself in the open

#neurodon list
aka
#neuroscience on #mastodon

https://docs.google.com/spreadsheets/d/1k3Q0LOewgIAJolbf-7XagK0twnpjeoyl2uLxR5U80vo/preview#gid=1993776511/html_view

Useful initiative by
@vineettiruvadi

#neurobuzz
#neurometa
#neuroscenius
#neuroIntro
#neuroscenius

Also follow tags & enable the 'Advanced Web Interface' if using browser ui:
https://mastodon.social/@teixi/109366102663397915

#Fediverse
#Newbieverse
#FediTips
#MastodonHelp
#TwitterMigration
#TwitterExodus

@neuroscience
@neuro
@cognition
@cogsci
@cogneurophys

elduvelle , 9 months ago to academicchatter

What’s the most difficult question you ever had in a job interview? @academicchatter
#workplace #Academia

teixi , 9 months ago

@elduvelle @academicchatter

my basic recommendation is that you have to afford for each one expert singularities while acknowledging the impossibility of total readiness.

then for those human emotional barriers no one can predict for.

learn on their trajectory, then imagine the ways how such research could possibly follow.

but then if you don't get a reply —usually or rarely— a sorry:
don't get down, take it as practice to improve next one.

final difficulty answering salary/future expects.