Timeline for Online estimation of covariance matrix
Current License: CC BY-SA 3.0
5 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Feb 11, 2014 at 18:31 | history | edited | Or Zuk | CC BY-SA 3.0 |
added 508 characters in body
|
| Feb 11, 2014 at 14:24 | comment | added | Ricardo Andrade | Dear @Or Zuk, thank you. I think it would be very helpful if you could copy the content of your comment into your answer. | |
| Feb 11, 2014 at 12:04 | comment | added | Or Zuk | The authors analyze the problem from an online learning perspective - that is, they give bounds on the regret: the gap between in performance between online estimation of mu and sigma and the optimal estimator when we have access to the entire data (the latter being the MLE, namely the sample mean and variance). They also provide an explicit update rule based on 'follow the leader' strategy for the mean and the covariance matrix when adding a new observation - see eq. (8) for the multivariate case. | |
| Feb 11, 2014 at 1:58 | comment | added | Ricardo Andrade | Dear @Or Zuk, could you please add some details to your answer? Thank you. | |
| Feb 11, 2014 at 0:11 | history | answered | Or Zuk | CC BY-SA 3.0 |