US 12,147,904 B2
Distance metrics and clustering in recurrent neural networks
Henry Markram, Pully (CH); Felix Schürmann, Grens (CH); Fabien Jonathan Delalondre, Geneva (CH); Ran Levi, Aberdeen (GB); Kathryn Hess Bellwald, Aigle (CH); and John Rahmon, Lausanne (CH)
Assigned to INAIT SA, Lausanne (CH)
Filed by INAIT SA, Lausanne (CH)
Filed on Feb. 13, 2023, as Appl. No. 18/167,958.
Application 18/167,958 is a continuation of application No. 16/710,176, filed on Dec. 11, 2019, granted, now 11,580,401.
Prior Publication US 2023/0351196 A1, Nov. 2, 2023
Int. Cl. G06K 9/62 (2022.01); G06F 18/22 (2023.01); G06F 18/23 (2023.01); G06N 3/082 (2023.01); G06N 3/088 (2023.01)
CPC G06N 3/088 (2013.01) [G06F 18/22 (2023.01); G06F 18/23 (2023.01); G06N 3/082 (2013.01)] 18 Claims
OG exemplary drawing
 
1. A method comprising:
determining a distance between first data and either second data or a reference using information that identifies topological patterns of activity that occur in response to an input of the first data into a recurrent neural network, wherein the distance is determined using a distance metric that treats a first subset of the topological patterns of activity differently from a second subset of the topological patterns of activity; and
clustering the first data based on the determined distances.