Pith. sign in

REVIEW 2 cited by

MQTransformer: Multi-Horizon Forecasts with Context Dependent and Feedback-Aware Attention

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2009.14799 v4 pith:T6V25GBH submitted 2020-09-30 cs.LG stat.ML

classification cs.LGstat.ML
keywords novelaccuracyattentionforecastforecastingimprovementsadvancescontext
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Recent advances in neural forecasting have produced major improvements in accuracy for probabilistic demand prediction. In this work, we propose novel improvements to the current state of the art by incorporating changes inspired by recent advances in Transformer architectures for Natural Language Processing. We develop a novel decoder-encoder attention for context-alignment, improving forecasting accuracy by allowing the network to study its own history based on the context for which it is producing a forecast. We also present a novel positional encoding that allows the neural network to learn context-dependent seasonality functions as well as arbitrary holiday distances. Finally we show that the current state of the art MQ-Forecaster (Wen et al., 2017) models display excess variability by failing to leverage previous errors in the forecast to improve accuracy. We propose a novel decoder-self attention scheme for forecasting that produces significant improvements in the excess variation of the forecast.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Structure-Informed Deep Reinforcement Learning for Inventory Management

    cs.LG 2025-07 conditional novelty 6.0 of 10

    A generic DirectBackprop deep RL policy, trained only on historical demand across many products, matches or beats classical inventory heuristics in five problem settings, and structural monotonicity penalties improve ...

  2. Forking-Sequences: Statistically and Computationally Efficient Multi-Horizon Forecasting with Reduced Volatility

    cs.LG 2025-10 conditional novelty 5.0 of 10

    Forking-sequences trains forecasting models on all forecast-creation dates jointly, reducing gradient and forecast variance relative to per-date window sampling.

Pith tools