Skip to content

Regularization fixes + switch for the pre-PR-15610 compatibility mode #15784

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
shahor02 wants to merge 2 commits into
base: dev
Choose a base branch
from
+187 −102
Open

Regularization fixes + switch for the pre-PR-15610 compatibility mode #15784

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
28d6213
DCAFitter: apply X-error regularization only where it is needed, drop…
shahor02 Sep 9, 2026
edb2171
Switch for pre PR-15610 compatibility mode
shahor02 Sep 10, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
18 changes: 14 additions & 4 deletions Common/DCAFitter/DCAFitterN_derivation.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -479,9 +479,11 @@ I_XX -> I_XX + 1/sigma_X,prior^2 .
In code this is implemented as

```cpp
constexpr float XRegErrFactor = 10.f;
const float sigmaX2 = C_YY * XRegErrFactor;
sxx += 1.f / sigmaX2;
static constexpr float XRegErrFactor = 10.f;
...
if (xRegErrFactor > 0.f) {
sxx += 1.f / (cyy * xRegErrFactor);
}
```

This is different from multiplying `I_XX` by a number below one. A reduction of
Expand All @@ -491,7 +493,15 @@ not alter `I_YY`, `I_YZ`, or `I_ZZ`.

The regularization should remain weak. It is a numerical stabilizer for badly
conditioned geometries, not an additional detector measurement of the local
track `X` coordinate.
track `X` coordinate. For this reason it is applied **only** where an invertible
single-track contribution is actually required, i.e. for the `I_i` entering the
chi2 minimization (`mTrcEInv`, hence `calcInverseWeight()`, `calcPCACoefs()`,
`calcChi2()` and the Newton Hessian). `calcPCACovMatrix()` rebuilds the `I_i`
with `TrackCovI::XRegNone`: there the prior is not needed (the sum over prongs is
inverted, not the individual terms, and a genuinely ill-conditioned sum is
detected and replaced by a loose dummy covariance), and including it would make
the reported longitudinal vertex error follow the dummy `XRegErrFactor * C_YY`
instead of the track slopes.

### `calcPCACovMatrix()`

Expand Down
Loading