Kristiaan Pelckmans

Kristiaan Pelckmans

Auto-generated profile, based on publicly available information

Kristiaan Pelckmans

Uppsala University

Research Focus & Metrics

Academic Career Length

18Years (2009-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
4 December 2020 in Science Advances
Verified
Journal Article
Open access
PDF file
AKAlexander Kotrschal
Alexander Kotrschal
ASAlexander Szorkovszky
Alexander Szorkovszky
JHJames Herbert-Read
James Herbert-Read
NBNatasha Bloch
Natasha Bloch
MRMaksym Romenskyy
Maksym Romenskyy
NKNiclas Kolm
Niclas Kolm
Collective motion occurs when individuals use social interaction rules to respond to the movements and positions of their neighbors. How readily these social decisions are shaped by selection remains unknown. Through artificial selection on fish (guppies, Poecilia reticulata) for increased group polarization, we demonstrate rapid evolution in how individuals use social interaction rules. Within only three generations, groups of polarization-selected females showed a 15% increase in polarization, coupled with increased cohesiveness, compared to fish from control lines. Although lines did not differ in their physical swimming ability or exploratory behavior, polarization-selected fish adopted faster speeds, particularly in social contexts, and showed stronger alignment and attraction responses to multiple neighbors. Our results reveal the social interaction rules that change when collective behavior evolves.

28 September 2017 in IEEE Transactions on Automatic Control
Verified
Journal Article
Closed access
SYSholeh Yasini
Sholeh Yasini
KPKristiaan Pelckmans
Kristiaan Pelckmans
In this paper, we study the prediction performance of the Kalman filter (KF) in a worst-case, minimax setting as studied in online machine learning, information - and game theory. The aim is to predict the sequence of observations almost as well as the best reference predictor (comparator) sequence in a comparison class. We prove worst-case bounds on the cumulative squared prediction errors using a priori knowledge about the complexity of reference predictor sequence. In fact, the performance of the KF is derived as a function of the performance of the best reference predictor and the total amount of drift that occurs in the schedule of the best comparator.

4 June 2011 in Machine Learning
Verified
Journal Article
Closed access
PDF file
Peter KarsmakersPeter Karsmakers
Peter Karsmakers
KPKristiaan Pelckmans
Kristiaan Pelckmans
KBKris De Brabanter
Kris De Brabanter
HHHugo Van hamme
Hugo Van hamme
JSJohan Suykens
Johan Suykens

6 October 2021
Verified
Book Chapter
Closed access
PDF file
ASAbel Souza
Abel Souza
KPKristiaan Pelckmans
Kristiaan Pelckmans
JTJohan Tordsson
Johan Tordsson

28 February 2020 in IEEE Intelligent Systems
Verified
Journal Article
Closed access
KPKristiaan Pelckmans
Kristiaan Pelckmans
Modern applications of FinTech are challenged by enormous volumes of financial data. One way to handle those is to adopt a streaming setting where data is only available to the algorithms during a very short time. When a new data point (financial transaction) is generated, it needs to be processed directly, and be forgotten immediately after. Especially, ongoing globalisation efforts in FinTech require modern methods of fault detection to be able to work efficiently through more than 10,000 financial transactions per second if they are to be deployed as a first line of defence. This paper investigates two algorithms able to perform well in this demanding setting: K-means and FADO. Especially, this manuscript provides supports for the claim that ‘the use of multiple clusters does not necessarily translate into increased detection performance’. To support this claim, results are reported when operating in a quasi-realistic case study of AML detection in realtime payment systems. We focus on two prototypical algorithms: the passive aggressive FADO assuming a single cluster, and the well-known K- means algorithm working with K > 1 clusters. We find - in this case - that the use of K-means with multiple clusters is unfavourable as (1) both tuning for K, as well as the need for additional complexity in the K- means algorithm challenges the computational constraints; (2) K-means introduces necessarily added variability (unreliability) in the results; (3) it requires dimensionality reduction, compromising interpretability of the detections; (4) the prevalence of singleton clusters adds unreliability to the outcome. This makes in the presented case FADO favourable over K-means (with K > 1).

1 September 2009 in IFAC Proceedings Volumes
Verified
Journal Article
Closed access
PDF file
KPKristiaan Pelckmans
Kristiaan Pelckmans
JSJohan Suykens
Johan Suykens