A metric-routing version of continual learning task ordering is NP-hard, and a Christofides-style polynomial algorithm achieves 3/2 and 3/2+r^{1-T} approximation ratios against an upper-bound objective.
Task difficulty awar e param- eter allocation & regularization for lifelong learning,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 1years
2024 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Algorithm Design for Continual Learning in IoT Networks
A metric-routing version of continual learning task ordering is NP-hard, and a Christofides-style polynomial algorithm achieves 3/2 and 3/2+r^{1-T} approximation ratios against an upper-bound objective.