Jump to content

Triplet loss: Revision history


For any version listed below, click on its date to view it. For more help, see Help:Page history and Help:Edit summary. (cur) = difference from current version, (prev) = difference from preceding version, m = minor edit, → = section edit, ← = automatic edit summary

(newest | oldest) View (newer 50 | ) (20 | 50 | 100 | 250 | 500)

6 July 2024

29 April 2024

15 November 2023

22 July 2023

9 June 2023

2 May 2023

25 February 2023

1 July 2022

31 May 2022

10 April 2022

4 April 2022

22 February 2022

24 January 2022

9 December 2021

29 September 2021

8 May 2021

6 May 2021

24 February 2021

  • curprev 12:4312:43, 24 February 202146.46.197.34 talk 6,608 bytes 0 This is just one of several early versions in machine learning, without going into all triplet and triangle equaities and inequalities in bordering fields. I don't know all variants, but it was at least known in late 1980. undo

8 February 2021

14 January 2021

25 October 2020

  • curprev 11:5111:51, 25 October 2020G.Obozinski talk contribs 6,699 bytes +579 Added a reference to the paper of Schultz and Joachims (2004) in which the triplet loss was introduced for the first time. The formulation of Schultz and Joachims is a SVM style formulation and their equation (8) is exactly equivalent to using a triplet loss to learn a Mahalanobis distance. undo

21 September 2020

25 July 2020

16 June 2020

14 June 2020

1 April 2020

19 September 2019

8 September 2019

2 September 2019

25 July 2019

30 June 2019

24 June 2019

(newest | oldest) View (newer 50 | ) (20 | 50 | 100 | 250 | 500)