Feedback mechanisms in sequence learning systems with temporal processing capability

A computing device comprising:

• a processor;
• and
• a memory operably coupled to the processor, the memory storing an instance of a first processing node, the first processing node comprising:
• a spatial pooler configured to receive input data that change over time, and generate a series of sparse distributed vectors over time based on the input data;
• and
• a sequence processor configured to:
• perform temporal processing on the series of sparse distributed vectors at the first processing node by at least generating sequence inputs varying over time based on the series of sparse distributed vectors,
• transmit sequence inputs via lateral connections within the first processing node,
• receive feedback inputs from a second processing node via first feedback connections,
• determine whether one or more learned temporal sequences of spatial patterns are included in the input data based on the transmitted sequence inputs and the received feedback inputs, and
• send output data to the second processing node for performing at least temporal processing on the output data, the output data indicating at least prediction or detection of one or more spatial patterns or one or more temporal sequences in the input data.

A non-transitory computer readable storage medium comprising instructions when executed by a processor cause the processor to instantiate a first processing node, the first processing node configured to:

• receive input data that change over time at the first processing node;
• generate a series of sparse distributed vectors over time based on the input data at the first processing node;
• perform temporal processing on the series of sparse distributed vectors at the first processing node by at least:
• generating sequence inputs varying over time based on the series of sparse distributed vectors,
• transmitting sequence inputs via lateral connections within the first processing node,
• receiving feedback inputs from a second processing node via first feedback connections, and
• determining whether one or more learned temporal sequences of spatial patterns are included in the input data based on the transmitted sequence inputs and the received feedback inputs;
• and
• send output data to the second processing node for performing at least temporal processing on the output data, the output data indicating at least prediction or detection of one or more spatial patterns or one or more temporal sequences in the input data.