Arrow Research search

Author name cluster

Yang Meng

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

6 papers
1 author row

Possible papers

6

EAAI Journal 2026 Journal Article

Research on multivariate time series prediction method for upper motion intention perception

  • Yang Meng
  • Shuhao Liang
  • Jinda Wang
  • Fei Niu
  • Wendong Wang
  • Zelin Ci

To address the limitations of relying on a single information source and the low accuracy in upper limb motion intention perception during exoskeleton-based rehabilitation training, a multivariate time-series prediction method that integrates a cross-graph convolution module with a stochastic synthetic attention mechanism is proposed. Specifically, a cross-graph convolution module based on Spatial Node Encoding (SNE) is developed to fuse data from the Inertial Measurement Unit (IMU) and visual signals, thereby capturing spatial relationships among variables. A multi - view topology mapping network with a stochastic synthetic attention mechanism is introduced to extract temporal features, and a Graph Convolutional Network - Long Short - Term Memory (GCN - LSTM) model is constructed. The proposed GCN - LSTM model is compared with the 1 - Dimensional Convolution - Long Short - Term Memory (1DConv - LSTM) and Bidirectional Long Short - Term Memory (Bi - LSTM) models through experiments. The results show that the GCN - LSTM achieves a joint trajectory fitting degree, R2, of 0. 9417. It represents an approximate 9 % improvement over 1DConv–LSTM and a 10 % improvement over Bi–LSTM, effectively enhancing the accuracy of upper limb motion intention perception and contributing to the improvement of rehabilitation training effects.

Details DOI

NeurIPS Conference 2025 Conference Paper

Transformers for Mixed-type Event Sequences

  • Felix Draxler
  • Yang Meng
  • Kai Nelson
  • Lukas Laskowski
  • Yibo Yang
  • Theofanis Karaletsos
  • Stephan Mandt

Event sequences appear widely in domains such as medicine, finance, and remote sensing, yet modeling them is challenging due to their heterogeneity: sequences often contain multiple event types with diverse structures—for example, electronic health records that mix discrete events like medical procedures with continuous lab measurements. Existing approaches either tokenize all entries, violating natural inductive biases, or ignore parts of the data to enforce a consistent structure. In this work, we propose a simple yet powerful Marked Temporal Point Process (MTPP) framework for modeling event sequences with flexible structure, using a single unified model. Our approach employs a single autoregressive transformer with discrete and continuous prediction heads, capable of modeling variable-length, mixed-type event sequences. The continuous head leverages an expressive normalizing flow to model continuous event attributes, avoiding the numerical integration required for inter-event times in most competing methods. Empirically, our model excels on both discrete-only and mixed-type sequences, improving prediction quality and enabling interpretable uncertainty quantification. We make our code public at https: //github. com/czi-ai/FlexTPP.

PDF Details

TMLR Journal 2024 Journal Article

Scaling Up Bayesian Neural Networks with Neural Networks

  • Zahra Moslemi
  • Yang Meng
  • Shiwei Lan
  • Babak Shahbaba

Bayesian Neural Networks (BNNs) offer a principled and natural framework for proper uncertainty quantification in the context of deep learning. They address the typical challenges associated with conventional deep learning methods, such as data insatiability, ad-hoc nature, and susceptibility to overfitting. However, their implementation typically either relies on Markov chain Monte Carlo (MCMC) methods, which are characterized by their computational intensity and inefficiency in a high-dimensional space, or variational inference methods, which tend to underestimate uncertainty. To address this issue, we propose a novel Calibration-Emulation-Sampling (CES) strategy to significantly enhance the computational efficiency of BNN. In this framework, during the initial calibration stage, we collect a small set of samples from the parameter space. These samples serve as training data for the emulator, which approximates the map between parameters and posterior probability. The trained emulator is then used for sampling from the posterior distribution at substantially higher speed compared to the standard BNN. Using simulated and real data, we demonstrate that our proposed method improves computational efficiency of BNN, while maintaining similar performance in terms of prediction accuracy and uncertainty quantification.

PDF Details

NeurIPS Conference 2024 Conference Paper

Unity by Diversity: Improved Representation Learning for Multimodal VAEs

  • Thomas M. Sutter
  • Yang Meng
  • Andrea Agostini
  • Daphné Chopard
  • Norbert Fortin
  • Julia E. Vogt
  • Babak Shahbaba
  • Stephan Mandt

Variational Autoencoders for multimodal data hold promise for many tasks in data analysis, such as representation learning, conditional generation, and imputation. Current architectures either share the encoder output, decoder input, or both across modalities to learn a shared representation. Such architectures impose hard constraints on the model. In this work, we show that a better latent representation can be obtained by replacing these hard constraints with a soft constraint. We propose a new mixture-of-experts prior, softly guiding each modality's latent representation towards a shared aggregate posterior. This approach results in a superior latent representation and allows each encoding to preserve information better from its uncompressed original features. In extensive experiments on multiple benchmark datasets and two challenging real-world datasets, we show improved learned latent representations and imputation of missing data modalities compared to existing methods.

PDF Details DOI