arXiv subjects

Jinghao Lu

Publications and source records attributed to Jinghao Lu.

3 recordsLinked to original sources

arXiv2026

EEO-TFV: Escape-Explore Optimizer for Web-Scale Time-Series Forecasting and Vision Analysis

Transformer-based foundation models have achieved remarkable progress in tasks such as time-series forecasting and image segmentation. However, they frequently suffer from error accumulation in multivariate long-sequence prediction and exhibit vulnerability to out-of-distribution samples in image-related tasks. Furthermore, these challenges become particularly pronounced in large-scale Web data analysis tasks, which typically involve complex temporal patterns and multimodal features. This complexity substantially increases optimization difficulty, rendering models prone to stagnation at saddle points within high-dimensional parameter spaces. To address these issues, we propose a lightweight Transformer architecture in conjunction with a novel Escape-Explore Optimizer (EEO). The optimizer enhances both exploration and generalization while effectively avoiding sharp minima and saddle-point traps. Experimental results show that, in representative Web data scenarios, our method achieves performance on par with state-of-the-art models across 11 time-series benchmark datasets and the Synapse medical image segmentation task. Moreover, it demonstrates superior generalization and stability, thereby validating its potential as a versatile cross-task foundation model for Web-scale data mining and analysis.

cs.LG↗

arXiv2025

IdealTSF: Can Non-Ideal Data Contribute to Enhancing the Performance of Time Series Forecasting Models?

Deep learning has shown strong performance in time series forecasting tasks. However, issues such as missing values and anomalies in sequential data hinder its further development in prediction tasks. Previous research has primarily focused on extracting feature information from sequence data or addressing these suboptimal data as positive samples for knowledge transfer. A more effective approach would be to leverage these non-ideal negative samples to enhance event prediction. In response, this study highlights the advantages of non-ideal negative samples and proposes the IdealTSF framework, which integrates both ideal positive and negative samples for time series forecasting. IdealTSF consists of three progressive steps: pretraining, training, and optimization. It first pretrains the model by extracting knowledge from negative sample data, then transforms the sequence data into ideal positive samples during training. Additionally, a negative optimization mechanism with adversarial disturbances is applied. Extensive experiments demonstrate that negative sample data unlocks significant potential within the basic attention architecture for time series forecasting. Therefore, IdealTSF is particularly well-suited for applications with noisy samples or low-quality data.

cs.LG↗

arXiv2017

Seeds Cleansing CNMF for Spatiotemporal Neural Signals Extraction of Miniscope Imaging Data

Miniscope calcium imaging is increasingly being used to monitor large populations of neuronal activities in freely behaving animals. However, due to the high background and low signal-to-noise ratio of the single-photon based imaging used in this technique, extraction of neural signals from the large numbers of imaged cells automatically has remained challenging. Here we describe a highly accurate framework for automatically identifying activated neurons and extracting calcium signals from the miniscope imaging data, seeds cleansing Constrained Nonnegative Matrix Factorization (sc-CNMF). This sc-CNMF extends the conventional CNMF with two new modules: i) a neural enhancing module to overcome miniscope-specific limitations, and ii) a seeds cleansing module combining LSTM to rigorously select and cleanse the set of seeds for detecting regions-of-interest. Our sc-CNMF yields highly stable and superior performance in analyzing miniscope calcium imaging data compared to existing methods.

q-bio.NC↗