Federated Foundation Models on Heterogeneous Time Series

Training a general-purpose time series foundation models with robust generalization capabilities across diverse applications from scratch is still an open challenge. Efforts are primarily focused on fusing cross-domain time series datasets to extract shared subsequences as tokens for training models on Transformer architecture. However, due to significant statistical heterogeneity across domains, this cross-domain fusing approach doesn't work effectively as the same as fusing texts and images. To tackle this challenge, this paper proposes a novel federated learning approach to address the heterogeneity in time series foundation models training, namely FFTS. Specifically, each data-holding organization is treated as an independent client in a collaborative learning framework with federated settings, and then many client-specific local models will be trained to preserve the unique characteristics per dataset. Moreover, a new regularization mechanism will be applied to both client-side and server-side, thus to align the shared knowledge across heterogeneous datasets from different domains. Extensive experiments on benchmark datasets demonstrate the effectiveness of the proposed federated learning approach. The newly learned time series foundation models achieve superior generalization capabilities on cross-domain time series analysis tasks, including forecasting, imputation, and anomaly detection.

Proactive Model Adaptation Against Concept Drift for Online Time Series Forecasting

Time series forecasting always faces the challenge of concept drift, where data distributions evolve over time, leading to a decline in forecast model performance. Existing solutions are based on online learning, which continually organize recent time series observations as new training samples and update model parameters according to the forecasting feedback on recent data. However, they overlook a critical issue: obtaining ground-truth future values of each sample should be delayed until after the forecast horizon. This delay creates a temporal gap between the training samples and the test sample. Our empirical analysis reveals that the gap can introduce concept drift, causing forecast models to adapt to outdated concepts. In this paper, we present \textsc{Proceed}, a novel proactive model adaptation framework for online time series forecasting. \textsc{Proceed} first operates by estimating the concept drift between the recently used training samples and the current test sample. It then employs an adaptation generator to efficiently translate the estimated drift into parameter adjustments, proactively adapting the model to the test sample. To enhance the generalization capability of the framework, \textsc{Proceed} is trained on synthetic diverse concept drifts. We conduct extensive experiments on five real-world datasets across various forecast models. The empirical study demonstrates that our proposed \textsc{Proceed} brings more performance improvements than the state-of-the-art online learning methods, significantly facilitating forecast models' resilience against concept drifts.

Spectral Differential Network Analysis for High-Dimensional Time Series

Spectral networks derived from multivariate time series data arise in many domains, from brain science to Earth science. Often, it is of interest to study how these networks change under different conditions. For instance, to better understand epilepsy, it would be interesting to capture the changes in the brain connectivity network as a patient experiences a seizure, using electroencephalography data. A common approach relies on estimating the networks in each condition and calculating their difference. Such estimates may behave poorly in high dimensions as the networks themselves may not be sparse in structure while their difference may be. We build upon this observation to develop an estimator of the difference in inverse spectral densities across two conditions. Using an L1 penalty on the difference, consistency is established by only requiring the difference to be sparse. We illustrate the method on synthetic data experiments, on experiments with electroencephalography data, and on experiments with optogentic stimulation and micro-electrocorticography data.

Advancing clinical trial outcomes using deep learning and predictive modelling: bridging precision medicine and patient-centered care

The integration of artificial intelligence [AI] into clinical trials has revolutionized the process of drug development and personalized medicine. Among these advancements, deep learning and predictive modelling have emerged as transformative tools for optimizing clinical trial design, patient recruitment, and real-time monitoring. This study explores the application of deep learning techniques, such as convolutional neural networks [CNNs] and transformerbased models, to stratify patients, forecast adverse events, and personalize treatment plans. Furthermore, predictive modelling approaches, including survival analysis and time-series forecasting, are employed to predict trial outcomes, enhancing efficiency and reducing trial failure rates. To address challenges in analysing unstructured clinical data, such as patient notes and trial protocols, natural language processing [NLP] techniques are utilized for extracting actionable insights. A custom dataset comprising structured patient demographics, genomic data, and unstructured text is curated for training and validating these models. Key metrics, including precision, recall, and F1 scores, are used to evaluate model performance, while trade-offs between accuracy and computational efficiency are examined to identify the optimal model for clinical deployment. This research underscores the potential of AI-driven methods to streamline clinical trial workflows, improve patient-centric outcomes, and reduce costs associated with trial inefficiencies. The findings provide a robust framework for integrating predictive analytics into precision medicine, paving the way for more adaptive and efficient clinical trials. By bridging the gap between technological innovation and real-world applications, this study contributes to advancing the role of AI in healthcare, particularly in fostering personalized care and improving overall trial success rates.

LMS-AutoTSF: Learnable Multi-Scale Decomposition and Integrated Autocorrelation for Time Series Forecasting

Time series forecasting is an important challenge with significant applications in areas such as weather prediction, stock market analysis, scientific simulations and industrial process analysis. In this work, we introduce LMS-AutoTSF, a novel time series forecasting architecture that incorporates autocorrelation while leveraging dual encoders operating at multiple scales. Unlike models that rely on predefined trend and seasonal components, LMS-AutoTSF employs two separate encoders per scale: one focusing on low-pass filtering to capture trends and the other utilizing high-pass filtering to model seasonal variations. These filters are learnable, allowing the model to dynamically adapt and isolate trend and seasonal components directly in the frequency domain. A key innovation in our approach is the integration of autocorrelation, achieved by computing lagged differences in time steps, which enables the model to capture dependencies across time more effectively. Each encoder processes the input through fully connected layers to handle temporal and channel interactions. By combining frequency-domain filtering, autocorrelation-based temporal modeling, and channel-wise transformations, LMS-AutoTSF not only accurately captures long-term dependencies and fine-grained patterns but also operates more efficiently compared to other state-of-the-art methods. Its lightweight design ensures faster processing while maintaining high precision in forecasting across diverse time horizons. The source code is publicly available at \url{http://github.com/mribrahim/LMS-TSF}

ChatTS: Aligning Time Series with LLMs via Synthetic Data for Enhanced Understanding and Reasoning

Understanding time series is crucial for its application in real-world scenarios. Recently, large language models (LLMs) have been increasingly applied to time series tasks, leveraging their strong language capabilities to enhance various applications. However, research on multimodal LLMs (MLLMs) for time series understanding and reasoning remains limited, primarily due to the scarcity of high-quality datasets that align time series with textual information. This paper introduces ChatTS, a novel MLLM designed for time series analysis. ChatTS treats time series as a modality, similar to how vision MLLMs process images, enabling it to perform both understanding and reasoning with time series. To address the scarcity of training data, we propose an attribute-based method for generating synthetic time series with detailed attribute descriptions. We further introduce Time Series Evol-Instruct, a novel approach that generates diverse time series Q&As, enhancing the model's reasoning capabilities. To the best of our knowledge, ChatTS is the first MLLM that takes multivariate time series as input, which is fine-tuned exclusively on synthetic datasets. We evaluate its performance using benchmark datasets with real-world data, including six alignment tasks and four reasoning tasks. Our results show that ChatTS significantly outperforms existing vision-based MLLMs (e.g., GPT-4o) and text/agent-based LLMs, achieving a 46.0% improvement in alignment tasks and a 25.8% improvement in reasoning tasks.

WinTSR: A Windowed Temporal Saliency Rescaling Method for Interpreting Time Series Deep Learning Models

Interpreting complex time series forecasting models is challenging due to the temporal dependencies between time steps and the dynamic relevance of input features over time. Existing interpretation methods are limited by focusing mostly on classification tasks, evaluating using custom baseline models instead of the latest time series models, using simple synthetic datasets, and requiring training another model. We introduce a novel interpretation method called Windowed Temporal Saliency Rescaling (WinTSR) addressing these limitations. WinTSR explicitly captures temporal dependencies among the past time steps and efficiently scales the feature importance with this time importance. We benchmark WinTSR against 10 recent interpretation techniques with 5 state-of-the-art deep-learning models of different architectures, including a time series foundation model. We use 3 real-world datasets for both time-series classification and regression. Our comprehensive analysis shows that WinTSR significantly outranks the other local interpretation methods in overall performance. Finally, we provide a novel and open-source framework to interpret the latest time series transformers and foundation models.

UTSD: Unified Time Series Diffusion Model

Transformer-based architectures have achieved unprecedented success in time series analysis. However, facing the challenge of across-domain modeling, existing studies utilize statistical prior as prompt engineering fails under the huge distribution shift among various domains. In this paper, a Unified Time Series Diffusion (UTSD) model is established for the first time to model the multi-domain probability distribution, utilizing the powerful probability distribution modeling ability of Diffusion. Unlike the autoregressive models that capture the conditional probabilities of the prediction horizon to the historical sequence, we use a diffusion denoising process to model the mixture distribution of the cross-domain data and generate the prediction sequence for the target domain directly utilizing conditional sampling. The proposed UTSD contains three pivotal designs: (1) The condition network captures the multi-scale fluctuation patterns from the observation sequence, which are utilized as context representations to guide the denoising network to generate the prediction sequence; (2) Adapter-based fine-tuning strategy, the multi-domain universal representation learned in the pretraining stage is utilized for downstream tasks in target domains; (3) The diffusion and denoising process on the actual sequence space, combined with the improved classifier free guidance as the conditional generation strategy, greatly improves the stability and accuracy of the downstream task. We conduct extensive experiments on mainstream benchmarks, and the pre-trained UTSD outperforms existing foundation models on all data domains, exhibiting superior zero-shot generalization ability. After training from scratch, UTSD achieves comparable performance against domain-specific proprietary models. The empirical results validate the potential of UTSD as a time series foundational model.

Leveraging Time-Series Foundation Model for Subsurface Well Logs Prediction and Anomaly Detection

The rise in energy demand highlights the importance of suitable subsurface storage, requiring detailed and accurate subsurface characterization often reliant on high-quality borehole well log data. However, obtaining complete well-log data is costly and time-consuming, with missing data being common due to borehole conditions or tool errors. While machine learning and deep learning algorithms have been implemented to address these issues, they often fail to capture the intricate, nonlinear relationships and long-term dependencies in complex well log sequences. Additionally, prior AI-driven models typically require retraining when introduced to new datasets and are constrained to deployment in the same basin. In this study, we explored and evaluated the potential of a time-series foundation model leveraging transformer architecture and a generative pre-trained approach for predicting and detecting anomalies in borehole well log data. Specifically, we fine-tuned and adopted the TimeGPT architecture to forecast key log responses and detect anomalies with high accuracy. Our proposed model demonstrated excellent performance, achieving R2 of up to 87% and a mean absolute percentage error (MAPE) as low as 1.95%. Additionally, the model's zero-shot capability successfully identified subtle yet critical anomalies, such as drilling hazards or unexpected geological formations, with an overall accuracy of 93%. The model represents a significant advancement in predictive accuracy and computational efficiency, enabling zero-shot inference through fine-tuning. Its application in well-log prediction enhances operational decision-making while reducing risks associated with subsurface exploration. These findings demonstrate the model's potential to transform well-log data analysis, particularly in complex geological settings.

Mixed Delay/Nondelay Embeddings Based Neuromorphic Computing with Patterned Nanomagnet Arrays

Patterned nanomagnet arrays (PNAs) have been shown to exhibit a strong geometrically frustrated dipole interaction. Some PNAs have also shown emergent domain wall dynamics. Previous works have demonstrated methods to physically probe these magnetization dynamics of PNAs to realize neuromorphic reservoir systems that exhibit chaotic dynamical behavior and high-dimensional nonlinearity. These PNA reservoir systems from prior works leverage echo state properties and linear/nonlinear short-term memory of component reservoir nodes to map and preserve the dynamical information of the input time-series data into nondelay spatial embeddings. Such mappings enable these PNA reservoir systems to imitate and predict/forecast the input time series data. However, these prior PNA reservoir systems are based solely on the nondelay spatial embeddings obtained at component reservoir nodes. As a result, they require a massive number of component reservoir nodes, or a very large spatial embedding (i.e., high-dimensional spatial embedding) per reservoir node, or both, to achieve acceptable imitation and prediction accuracy. These requirements reduce the practical feasibility of such PNA reservoir systems. To address this shortcoming, we present a mixed delay/nondelay embeddings-based PNA reservoir system. Our system uses a single PNA reservoir node with the ability to obtain a mixture of delay/nondelay embeddings of the dynamical information of the time-series data applied at the input of a single PNA reservoir node. Our analysis shows that when these mixed delay/nondelay embeddings are used to train a perceptron at the output layer, our reservoir system outperforms existing PNA-based reservoir systems for the imitation of NARMA 2, NARMA 5, NARMA 7, and NARMA 10 time series data, and for the short-term and long-term prediction of the Mackey Glass time series data.