Messages
Dear all,
Next week we will have two papers presented.
Berit will present Efron (1979): Bootstrap Methods: Another Look at the Jackknife. AoS.
Michaela will present Stein (1981): Estimation of the Mean of a Multivariate Normal Distribution. AoS.
See you all then!
Dear all,
Next week we will have two papers presented.
Berit will present Efron (1979): Bootstrap Methods: Another Look at the Jackknife. AoS.
Jonas will present Ferguson (1973): A Bayesian Analysis of Some Nonparametric Problems. AoS.
See you all then!
Dear all,
Next week we will have two papers presented.
- Borgar will present I.J.Good (1952): Rational Decisions. JRSSB
- Michaela will present Hyvärinen (2005): Estimation of Non-Normalized Statistical Models by Score Matching. JMLR
Looking forward!
Dear all,
On 22 October we will have two presentations:
- Jonas will present Fraley & Raftery (2002): Model-Based Clustering, Discriminant Analysis, and Density Estimation. JASA.
- Espen will present Liu et al (2021): Minimax rates in sparse, high-dimensional change point detection. AoS.
Dear all,
On 15 October Michaela will present
Stephens (2017): False discovery rates: a new deal
See you then.
Dear students,
All students will be away next week. After long and careful consideration, we have concluded that having a class without any students would be a bit difficult.
The planned session 8 October is therefore canceled.
PA
Dear all,
Next week Berit will present
Angrist, Imbens og Rubin (1996): Identification of Causal Effects Using Instrumental Variables
See you then!
Dear all,
Next week Borgar will present
Blei et al (2017), "Variational Inference: A Review for Statisticians"
See you then!
Dear all,
Next week Espen will present
Wilson (2025) Deep Learning is Not So Mysterious or Different
See you then!
Dear all,
Next week Jonas will present
See you then!
Dear all,
Next week we will discuss the following two papers
Breiman (2001) Statistical Modeling: The Two Cultures (presented by Borgar)
Ioannidis (2005) Why Most Published Research Findings Are False (presented by Berit)
Looking forward to interesting discussions!
Leiv & Per August
This is a somewhat different course than the usual ones at the university: rather than traditional lectures, candidates will take turns presenting papers to the class.
Each candidate will present two papers over the course of the semester:
One hallmark/classic paper in statistics or machine learning, chosen by us.
One paper related to their own research. For this one, we'll ask you to suggest 2–3 candidate papers for us to choose from. We recommend that you ask your supervisor to suggest some papers.
Each presentation will run for about 30 minutes, followed by a 15-minute discussion, with two papers presented each class. Everyone is expected to have read the papers in advance and to contribute to the discussions afterward.
We aim to make the course interesting and relevant for both statisticians and machine learners, with a good mix of topics from both disciplines throughout the semester.
The course ends ...