|
|
|
|
Safety_Critical_Machine_Learning |
|
|
|
ANAM (Agnostic Neurons Array Memeory) is the result of a research started to give answers to some issues raised in the DARPA HyDDENN (Hyper-Dimensional Data Enabled Neural Networks) program. |
ANAM™ (Agnostic Neurons Array Memory™) is an algorithm which
allows the associative learning of numerical vectors through agnostic prototypes
analyzed in rank order coding. The ANAM algorithm uses an array of
"agnostic" neurons that have not learned and do not learn information
from data. The ordering of the agnostic neurons in the interaction with the
input data produces indexes that will be used to address the memory. Memory is
used to store the desired outputs in the learning phase and to retrieve the
same in the inference phase. The great criticality of this algorithm is the
perfect synthesis of generalization and selectivity capabilities. The
preprocessing of the input data is the determining factor for the correct
functioning of the algorithm and must be evaluated on a case-by-case basis.
|
|
|
General Synaptics Aerospace_and_Defence_Machine_Learning_Company VAT NUMBER:_IT02670700992 REA NUMBER: GE-503104 Email:_luca.marchese@synaptics.org |
|
|
|
Copyright© 2026 General Synaptics |
