TimeMCL uses Winner-Takes-All training of multiple heads to quantize the conditional distribution of future time series, producing diverse forecasts at low inference cost.
Inference.We used the official experimental protocol for evaluation in this benchmark (e.g.,(Rasul et al., 2021a))
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Winner-takes-all for Multivariate Probabilistic Time Series Forecasting
TimeMCL uses Winner-Takes-All training of multiple heads to quantize the conditional distribution of future time series, producing diverse forecasts at low inference cost.