US 11,853,884 B2
Many or one detection classification systems and methods
Saeed Mosayyebpour Kaskari, Irvine, CA (US)
Assigned to Synaptics Incorporated, San Jose, CA (US)
Filed by SYNAPTICS INCORPORATED, San Jose, CA (US)
Filed on Apr. 28, 2021, as Appl. No. 17/243,519.
Application 17/243,519 is a continuation in part of application No. 16/724,025, filed on Dec. 20, 2019, granted, now 11,080,600.
Application 16/724,025 is a continuation in part of application No. 15/894,883, filed on Feb. 12, 2018, granted, now 10,762,891, issued on Sep. 1, 2020.
Claims priority of provisional application 62/465,723, filed on Mar. 1, 2017.
Claims priority of provisional application 62/457,663, filed on Feb. 10, 2017.
Prior Publication US 2021/0248470 A1, Aug. 12, 2021
Int. Cl. G06N 3/08 (2023.01); G10L 25/51 (2013.01); G10L 25/30 (2013.01); G10L 15/16 (2006.01); G10L 15/22 (2006.01); G10L 15/06 (2013.01); G10L 15/08 (2006.01)
CPC G06N 3/08 (2013.01) [G10L 15/063 (2013.01); G10L 15/16 (2013.01); G10L 15/22 (2013.01); G10L 25/30 (2013.01); G10L 25/51 (2013.01); G10L 2015/088 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method for training an event classifier comprising:
receiving a training dataset including a plurality of input samples having segmented labeled data;
computing a classifier output for each input sample of the training dataset in a forward pass through the classifier;
tuning a many-or-one detection (MOOD) cost function using a hyperparameter in accordance with at least one event classifier goal;
updating weights and biases of the classifier through a backward pass, including determining whether an input frame is in a Region of Target (ROT), and estimating the update of the weights and the biases of the classifier;
wherein the classifier is trained using the tuned MOOD cost function to cause the classifier to spike at least one time during a duration of the event.