arXiv subjects

Jiazhen Zhu

Publications and source records attributed to Jiazhen Zhu.

3 recordsLinked to original sources

arXiv2025

MASE: Interpretable NLP Models via Model-Agnostic Saliency Estimation

Deep neural networks (DNNs) have made significant strides in Natural Language Processing (NLP), yet their interpretability remains elusive, particularly when evaluating their intricate decision-making processes. Traditional methods often rely on post-hoc interpretations, such as saliency maps or feature visualization, which might not be directly applicable to the discrete nature of word data in NLP. Addressing this, we introduce the Model-agnostic Saliency Estimation (MASE) framework. MASE offers local explanations for text-based predictive models without necessitating in-depth knowledge of a model's internal architecture. By leveraging Normalized Linear Gaussian Perturbations (NLGP) on the embedding layer instead of raw word inputs, MASE efficiently estimates input saliency. Our results indicate MASE's superiority over other model-agnostic interpretation methods, especially in terms of Delta Accuracy, positioning it as a promising tool for elucidating the operations of text-based models in NLP.

cs.CL↗

arXiv2020

Dynamics of a diffusive competitive model on a periodically evolving domain

In this paper, we are concerned with a two-species competitive model with diffusive terms on a periodically evolving domain and study the impact of the spatial periodic evolution on the dynamics of the model. The Lagrangian transformation approach is adopted to convert the model from a changing domain to a fixed one with the assumption that the evolution of habitat is uniform and isotropic. The ecological reproduction indexes of the linearized model are given as thresholds to reveal the dynamic behaviour of the competitive model by discussing the relations between the initial boundary value problem and its corresponding periodic problem. Our theoretical results show that a lager evolving rate benefits the persistence of competitive populations for both sides in the long run. Numerical experiments illustrate that two competitive species, one of which survive and the other vanish on a fixed domain, both survive on a domain with a large evolving rate and both vanish on a domain with a small evolving rate.

math.AP↗

arXiv2018

Coordinating Disaster Emergency Response with Heuristic Reinforcement Learning

A crucial and time-sensitive task when any disaster occurs is to rescue victims and distribute resources to the right groups and locations. This task is challenging in populated urban areas, due to the huge burst of help requests generated in a very short period. To improve the efficiency of the emergency response in the immediate aftermath of a disaster, we propose a heuristic multi-agent reinforcement learning scheduling algorithm, named as ResQ, which can effectively schedule the rapid deployment of volunteers to rescue victims in dynamic settings. The core concept is to quickly identify victims and volunteers from social network data and then schedule rescue parties with an adaptive learning algorithm. This framework performs two key functions: 1) identify trapped victims and rescue volunteers, and 2) optimize the volunteers' rescue strategy in a complex time-sensitive environment. The proposed ResQ algorithm can speed up the training processes through a heuristic function which reduces the state-action space by identifying the set of particular actions over others. Experimental results showed that the proposed heuristic multi-agent reinforcement learning based scheduling outperforms several state-of-art methods, in terms of both reward rate and response times.

cs.LG↗