CORTEX Apps
Semantia 1.4
Semantia combines features of cascading,associative and symbolic (hybrid) artificial neural networks and isthe central associative processing module of CORTEX artificialintelligence systems. This comprehensive framework aims to provideresearchers in the fields of AI, linguistics and animalcommunication with a powerful autonomous tool that simultaneouslyrecognizes and associates patterns from multiple parallel stimuli.Semantia automatically generates a neural network based onobserved conversations. All associations are saved and built uponeach time the app is opened unless the button "Clear Data" isclicked.Semantia is like a child with no knowledge -- she learnseverything she knows by interacting with the user. As her knowledgebase grows, her statements become more coherent and unexpected.Enabling Conversation Mode allows Semantia to providesyntax-free, child-like speech and text feedback triggered byspeech input (Click "Listen" or enable Continuous Listening)Enabling Continuous Listening allows Semantia to learn by"eavesdropping" on any speech input -- from the user, radio,television, etc.If Semantia is left unattended, Conversation Mode should bedisabled to maximize learning rate.Use voice command "close" to automatically generate a text filecontaining neural weights, connections and activations (array.txtin top SD directory) which can be used to generate neural networksfor conversational AI or imported into a network analysis tool suchas NodeXL.
Emergent X 1.0
This app allows the user to interactively experiment with andvisualize the training of a single neuron (feature detector) in anunsupervised neural network. The user provides the online "trainingdata" via taking photos. Instructions: take photos and be patient-- even small neural networks are computationally intensive! Thenetwork converges on the first "principal component" or most oftenoccurring 2 x 2 feature pattern that multiple images share incommon. This network implements only local learning rules: 1. Aneuron is competitive: if it frequently does not fire, it decreasesits threshold to respond to stimuli with greater variance. 2. Aneuron is thrifty: if it frequently fires, it increases itsthreshold 3. When a neuron fires, it tunes its "weights" to move ina direction which will increase the likelihood of a similar patterntriggering it in the future. Example: if a post synaptic neuron wasresponding to a high amplitude signal, but now responds to a loweramplitude signal (due to threshold decrease), its weights will movetowards the lower amplitude signal. Network structure: 1. Inputlayer: 8 x 8 x 3 (3 color channels) 2. Convolutional layer: 2 x 2 x3 ( 1 neuron with 12 synapses) 3. Activation Map: 7 x 7 x 1 Note:Color channels are preprocessed to respond similarly regardless oflighting.
EEG Profile 1.3
Similar in concept to a neural " fingerprint ", a profile ofaverage EEG frequencies, particularly resting state EEG, isfrequently correlated with various mental states as well asneurological conditions such as autism and schizophrenia. UtilizesNeurosky MindWave Mobile to record and view real-time runningaverages of EEG data (sample number length adjustable from 5 to200). Optional voice notification when EEG deviates from benchmarkaverage (first n-length samples) with adjustable similarity.
MagScan Magnetic Field Scanner 1.3
To work correctly, Android device must have built-in gyroscope,accelerometer, and 3-axis magnetometer. Also, auto-rotate ofdisplay must be disabled in settings. To begin using, hold phoneflat with the top pointing away from you and slowly move back andforth horizontally. Tilt phone up a small amount and rotate backand forth on the new imaginary plane. Continue tilting phone upafter the completion of each arc. Phone must not tilt upwards morethan 90 degrees. For more accurate results, phone can be mounted inmechanized harness. Default pixel size is 15 -- can adjust withslider from pixel size of 1 to 30. Colorization options: Direction+ Strength: Field direction is indicated by colors: Red is north,blue is south, green is a perpendicular field. Field strength(relative to max) is indicated by pixel brightness. Polarity +Strength: Field polarity is indicated by colors: Red is north, blueis south. Field strength (relative to max) is indicated by pixelbrightness. Strength: Field polarity is not indicated. Red is highfield strength, blue is low field strength relative to max. How itworks: As the phone is moved, the app uses the phone's built-in3-axis magnetometer to record the field strength on each of 3 axis.
