Evolved Machines, Inc.

Paul A Rhodes

Brian Taba

A first array of simulated neurons having trees of output branches and a second array of simulated neurons having trees of input branches are generated. Thereafter, the output branches of one or more of the simulated neurons of the first array and the input branches of one or more of the simulated neurons of the second array are grown and connections are formed between individual output branches of the simulated neurons of the first array and individual input branches of the simulated neurons of the second array that grow to within a vicinity of each other.

Growing simulated biological neural circuits in a simulated physical volume

Simulated neural circuitry is trained using sequences of images representing moving objects, such that the simulated neural circuitry recognizes objects by having the presence of lower level object features that occurred in temporal sequence in the images representing moving objects trigger the activation of higher level object representations. Thereafter, an image of an object that includes lower level object features is received, the trained simulated neural circuitry activates a higher level representation of the object in response to the lower level object features from the image, and the object is recognized using the trained simulated neural circuitry.

Invariant object recognition

A simulated neural circuit includes a plurality of simulated neurons. The simulated neurons have input branches that are configured to connect to a plurality of inputs and activate in response to activity in the inputs to which they are connected. In addition, the simulated neurons are configured to activate in response to activity in their input branches. Initial connections are formed between various input branches and various inputs and a set of the inputs are activated. Thereafter, the stability of connections between input branches and inputs to which they are connected is moderated based on the activated set of inputs and a pattern of activity generated in the input branches and simulated neurons in response to the activated set of inputs.

Activity-dependent generation of simulated neural circuits

A system and method of identifying the computing architecture used by the mammalian visual system and to implement it in simulations and software algorithms, and in hardware components, is described.

A simulated neural element includes a cell body and one or more simulated branches. Simulated branches are configured to receive input signals and to activate when a combination of the signals received during a specified window of time exceeds a branch activation threshold level. The simulated cell body is configured to activate when a combination of activity in the simulated branches during another specified window of time exceeds a cell body activation threshold level. The branch and cell body activation threshold levels may be automatically and locally regulated so that the actual branch activation rates for the simulated branches approximate desired branch activation rates and the actual cell body activation rate for the simulated cell body approximates a desired cell body activation rate. Such 'homeostatic' regulation of branch and cell firing thresholds, done locally (i.e. individually for each branch and cell), may enhance the performance of artificial neural circuitry.

Regulating activation threshold levels in a simulated neural circuit

A three-dimensional voxel space is generated in which to generate a simulated neural circuit. The voxel space includes a plurality of voxels that store localized information. After the voxel space is generated, a plurality of simulated branched neurons, each of which has one or more input and/or output branches that occupy at least one of the voxels, are embedded in the voxel space. One or more of the branches of the plurality of simulated neurons then are generated in a manner that changes the voxels occupied by the grown branches, and the localized information stored in the voxels is updated to reflect the changes in the voxels occupied by the grown branches.

Generating simulated neural circuits in a voxel space

A method for the orthogonalization of initially overlapping vectors, the uses for which include the re-encoding and decoding of representations triggered by sensory arrays.

Growing community of inventors

Palm Beach, FL, United States of America

Paul A Rhodes