mlsanityData to 14 Oct 2026

Edge Selection for the Effective use of Piecewise-Constant Distributions as Neural Network Outputs for Event Prediction

arXiv 2507.21616 ↗First posted 29 Jul 2025cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

We study the output representation of a neural network used for next event prediction. We propose partitioning the time axis into a fixed set of intervals and having a neural network output a categorical distribution over them, which we map to a (mostly) piecewise-constant probability density. We present an optimization procedure that selects interval edges in order to maximize data likelihood under the representation. The representation is well suited to processes whose inter-event distribution is a mixture of smooth and sharply peaked componentsunicodex2013a pattern we find common in event data recorded from real-world processes.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 16 of its 37 references that are in the map. First posted on arXiv 29 Jul 2025.

Builds on

16 of 37 references.

References · 16

  1. 2014Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translationcited by 4,057
  2. 2015Cyclical Learning Rates for Training Neural Networkscited by 629
  3. 2019Predicting Dynamic Embedding Trajectory in Temporal Interaction Networkscited by 448
  4. 2020Transformer Hawkes Processcited by 235
  5. 2019Intensity-Free Learning of Temporal Point Processescited by 140
  6. 2019Fully Neural Network based Model for General Temporal Point Processescited by 122
  7. 2021Neural Temporal Point Processes: A Reviewcited by 114
  8. 2019Continuous and Discrete-Time Survival Prediction with Neural Networkscited by 70
  9. 2023EasyTPP: Towards Open Benchmarking Temporal Point Processescited by 68
  10. 2022HYPRO: A Hybridly Normalized Probabilistic Model for Long-Horizon Prediction of Event Sequencescited by 57
  11. 2018Learning Temporal Point Processes via Reinforcement Learningcited by 40
  12. 2018Deep Reinforcement Learning of Marked Temporal Point Processescited by 33
4 more
  1. 2023Add and Thin: Diffusion for Temporal Point Processescited by 23
  2. 2024Decomposable Transformer Point Processescited by 16
  3. 2023On the Predictive Accuracy of Neural Temporal Point Process Models for Continuous-time Event Datacited by 14
  4. 2025Random irregular histogramscited by 1

Built on by

Not cited yet.