commit 1fe5a09e71f7f073f996ec4a95aac5353db01ef3
parent 7d7d34668c88759880bc8526bbd8319ff10669c3
Author: Silas Brack <silasbrack@gmail.com>
Date: Tue, 7 Apr 2026 15:39:03 +0200
fix: various styling fixes
Diffstat:
| M | main.typ | | | 66 | +++++++++++++++++++++++++++++++++--------------------------------- |
1 file changed, 33 insertions(+), 33 deletions(-)
diff --git a/main.typ b/main.typ
@@ -5,7 +5,7 @@
y: 1.0cm,
),
)
-#set text(size: 12pt, font: "Cooper Hewitt")
+#set text(size: 12pt, font: "PT Sans")
#set par(
first-line-indent: 0pt,
spacing: 2.0em,
@@ -69,7 +69,7 @@
// - Independent
-#line(length: 10%, stroke: 2pt)
+#line(length: 7%, stroke: 2pt)
== EXPERIENCE
#grid(
@@ -162,31 +162,33 @@
"Effortless Bayesian Deep Learning: Tapping Into the Potential of Modern Optimizers," with Søren Hauberg.
]
)
- // #grid(
- // rows: (auto, auto),
- // row-gutter: 0.8em,
- // [
- // #grid(
- // columns: (
- // 3.5fr,
- // 1.5fr,
- // ),
- // align(left)[
- // *B.Sc. General Engineering* \
- // _Technical University of Denmark_
- // ],
- // align(right)[
- // *Sep 2017 -- Aug 2020* \
- // _Copenhagen, Denmark_
- // ],
- // )
- // ],
- // text(size: 10pt)[
- // // - Specialisation in _cyber materials_.
- // - Thesis: "Membrane Capacitive Desalination," with Johan Hjelm
- // ]
- // )
- ]
+ ],
+ // [2017--2020],
+ // [
+ // #grid(
+ // rows: (auto, auto),
+ // row-gutter: 0.8em,
+ // [
+ // #grid(
+ // columns: (
+ // 1fr,
+ // auto,
+ // ),
+ // align(left)[
+ // *B.Sc. General Engineering* \
+ //
+ // ],
+ // align(right)[
+ // _Technical University of Denmark_
+ // ],
+ // )
+ // ],
+ // text(size: 10pt)[
+ // // - Specialisation in _cyber materials_.
+ // "Membrane Capacitive Desalination," with Johan Hjelm
+ // ]
+ // )
+ // ],
)
#line(length: 7%, stroke: 2pt)
@@ -212,13 +214,12 @@
*Marginal Likelihood Training of Linearized Laplace Approximations Without Hessian Reductions*
],
align(right)[
- _DTU Compute_ \
- w/ Søren Hauberg
+ _DTU Compute_
],
)
],
text(size: 10pt)[
- Developed a novel method for computing the Laplace approximation using only Jacobian-vector products in JAX, implementing posterior sampling (for inference) and the log-determinant (for optimising the marginal likelihood during training) of the Laplace covariance without explicitly instantiating it.
+ Developed a novel method for computing the Laplace approximation using only Jacobian-vector products in JAX, implementing posterior sampling (inference) and the log-determinant (optimising the marginal likelihood during training) of the Laplace covariance without explicitly instantiating it. Supervised by Søren Hauberg.
]
)
],
@@ -238,13 +239,12 @@
*Bayesian Metric Learning for Uncertainty Quantification in Image Retrieval*
],
align(right)[
- _DTU Compute_ \
- w/ Søren Hauberg
+ _DTU Compute_
],
)
],
text(size: 10pt)[
- Developed a method for training Bayesian neural networks in metric learning and demonstrate its effectiveness on small- to large-scale image datasets in yielding well-calibrated uncertainty estimates. We proved that contrastive loss constitutes a valid log-likelihood in spherical space and present a novel decomposition of its Generalized Gauss-Newton (GGN) approximation. Our paper was accepted to NeurIPS 2023 @warburg2023bayesian in New Orleans, which I had the fortune of attending.
+ Developed a method for training Bayesian neural networks in metric learning and demonstrate its effectiveness on small- to large-scale image datasets in yielding well-calibrated uncertainty estimates. We proved that contrastive loss constitutes a valid log-likelihood in spherical space and present a novel decomposition of its Generalized Gauss-Newton (GGN) approximation. Our paper was accepted to NeurIPS 2023 @warburg2023bayesian in New Orleans, which I had the fortune of attending. Supervised by Søren Hauberg.
]
)
],
