Self-supervised learning for accelerometer-based human activity recognition: A survey

A Logacjov - Proceedings of the ACM on Interactive, Mobile …, 2024‏ - dl.acm.org
Self-supervised learning (SSL) has emerged as a promising alternative to purely supervised
learning, since it can learn from labeled and unlabeled data using a pre-train-then-fine-tune …

[HTML][HTML] Probabilistic forecasting of renewable energy and electricity demand using Graph-based Denoising Diffusion Probabilistic Model

A Miraki, P Parviainen, R Arghandeh - Energy and AI, 2025‏ - Elsevier
Renewable energy production and the balance between production and demand have
become increasingly crucial in modern power systems, necessitating accurate forecasting …

Efficient time series processing for transformers and state-space models through token merging

L Götz, M Kollovieh, S Günnemann… - arxiv preprint arxiv …, 2024‏ - arxiv.org
Transformer architectures have shown promising results in time series processing. However,
despite recent advances in subquadratic attention mechanisms or state-space models …

Enhancing Hierarchical Sales Forecasting with Promotional Data: A Comparative Study Using ARIMA and Deep Neural Networks.

M Teixeira, JM Oliveira… - Machine Learning & …, 2024‏ - search.ebscohost.com
Retailers depend on accurate sales forecasts to effectively plan operations and manage
supply chains. These forecasts are needed across various levels of aggregation, making …

Causal Mechanism-Enabled Zero-Label Learning for Power Generation Forecasting of Newly-Built PV Sites

P Zhao, W Hu, D Cao, R Huang, X Wu… - IEEE Transactions …, 2024‏ - ieeexplore.ieee.org
Power forecasting of newly built photovoltaic (PV) sites faces huge challenges owing to the
lack of sufficient training samples. To this end, this paper proposes an unsupervised zero …

Foundation Models for CPS-IoT: Opportunities and Challenges

O Baris, Y Chen, G Dong, L Han, T Kimura… - arxiv preprint arxiv …, 2025‏ - arxiv.org
Methods from machine learning (ML) have transformed the implementation of Perception-
Cognition-Communication-Action loops in Cyber-Physical Systems (CPS) and the Internet of …

[HTML][HTML] Exploration of Foundational Models for Blood Glucose Forecasting in Type-1 Diabetes Pediatric Patients

S Rancati, P Bosoni, R Schiaffini, A Deodati… - Diabetology, 2024‏ - mdpi.com
Aims: The accurate prediction of blood glucose (BG) levels is critical for managing Type-1
Diabetes (T1D) in pediatric patients, where variability due to factors like physical activity and …

Are Time Series Foundation Models Ready to Revolutionize Predictive Building Analytics?

OB Mulayim, P Quan, L Han, X Ouyang… - Proceedings of the 11th …, 2024‏ - dl.acm.org
Recent advancements in large language models have spurred significant developments in
Time Series Foundation Models (TSFMs). These models claim great promise in performing …

Introducing ProsperNN—a Python package for forecasting with neural networks

N Beck, J Schemm, C Ehrig, B Sonnleitner… - PeerJ Computer …, 2024‏ - peerj.com
We present the package prosper_nn, that provides four neural network architectures
dedicated to time series forecasting, implemented in PyTorch. In addition, prosper_nn …

Contextualized Sequence Likelihood: Enhanced Confidence Scores for Natural Language Generation

Z Lin, S Trivedi, J Sun - arxiv preprint arxiv:2406.01806, 2024‏ - arxiv.org
The advent of large language models (LLMs) has dramatically advanced the state-of-the-art
in numerous natural language generation tasks. For LLMs to be applied reliably, it is …