All Tags
AWS
ai
algorithm-design
architecture
browser
cloud
cloud-efficiency
cloud-principles
cost-reduction
data-centric
data-compression
data-processing
deployment
design
documentation
edge-computing
email-sharing
energy-efficiency
energy-footprint
enterprise-optimization
green-ai
hardware
libraries
llm
locality
machine-learning
maintainability
management
measured
microservices
migration
mobile
model-optimization
model-training
multi-objective
network-traffic
parameter-tuning
performance
queries
rebuilding
scaling
services
storage-optimization
strategies
tabs
template
testing
workloads
Tactic: Adaptive Ensemble
Tactic sort:
Awesome Tactic
Type: Architectural Tactic
Category: resource-adaptation
Title
Adaptive Ensemble
Description
Adaptive ensemble aggregates the predictions of multiple models to adapt to concept drift (CD). Based on detection or on periodic training, models are trained on different slices of the data stream and dynamically weighting the contribution of each model based on recent prediction performance. It is a more general approach that can adequately handle cases of gradual, abrupt, and reoccurring CD [22], but is potentially less effective compared to approaches that specifically target a particular type of CD
Participant
Machine Learning Practitioner.
Related software artifact
Machine Learning 'Artefact'.
Context
Concept Drift. Architectural Design Decisions. Evolvability.
Software feature
Regular Re-Training. Dynamic Model Weighing. Model Aggregation.
Tactic intent
Reduce concept drift.
Target quality attribute
Concept drift.
Other related quality attributes
Evolvability.
Measured impact
< unknown >
Source
Evolvability of Machine Learning-based Systems: An Architectural Design Decision Framework by Joran Leest et al. (DOI: 10.1109/ICSA-C57050.2023.00033)Graphical representation
