CascadiaPrime

 Home    About    Blog    Future of Life Institute    Oxford FHI    Cambridge Center for Existential Risk   Machine Intelligence Research Institute     US Brain Project    EU Brain Project     China Brain Project  

 Blue Brain Project    EU Cognition Network   AAAI    CAIAC     Allen Institute for AI     AI for the Brain     AI 100     Center for Brains, Minds & Machines    Lifeboat     Long Now     PMD     AI International  

 
 


CascadiaPrime Cognition - Multimodal Perception


Human perception involves multimodal sensory integration - sight, sound, smell, touch, taste - which, depending upon circumstance, allows us to make meaningful representations of the world around us. Each sensory mechanism comes at an energy cost. It is therefore expressed to greater or lesser degree - or not at all, defined by a species or individual's ecological niche.

Senses are innate, but clearly involve learning mechanisms. Attention is part of the energy rationing system - chemical (glucose) and the computational rationing system (combinatorial explosion management).

Reverse engineering human multimodal sensory integration may be useful in localizing the "Master Algorithm". It may also create superintelligence risks that might be mitigated by only "partial deployment" of capabilities.

    
  Chaos Paper: Sequential memory: Binding dynamics (October 2015)
  
  Wired: Electronic skin will change the way we interact with tech (October 2015)
  
  ACM TechNews: How Sensorimotor Intelligence May Develop (October 2015)
  
  New understanding of the mechanics of learning - “synaptic eligibility traces” effectively tag the synapses activated by the stimuli (October 2015)
  
  The Power of Two: Neural Cliques Provide a Prewired Framework For How Neurons Connect (October 2015)
  
  IEEE Paper: Active Perception for Multimodal Object Category Recognition Using Information Gain (October 2015)
  
  Ruslan Salakhutdinov - Deep learning - Changing the playing field of artificial intelligence (CIFAR)(July 2015)
  
  Andrew Ng Deep Learning (May 2013)
  
  Structural and functional connectivity of visual and auditory attentional networks: insights from the Human Connectome Project (September 2015)
  
  Numenta working on sensorimotor learning and prediction (SLAP )(October 2015)
  
  New insight into light detection in vertebrates - light detection outside the eyes for multiple purposes (October 2015)
  
Top