arXiv subjects

Nengfeng Zhou

Publications and source records attributed to Nengfeng Zhou.

9 recordsLinked to original sources

arXiv2024

Towards a framework on tabular synthetic data generation: a minimalist approach: theory, use cases, and limitations

We propose and study a minimalist approach towards synthetic tabular data generation. The model consists of a minimalistic unsupervised SparsePCA encoder (with contingent clustering step or log transformation to handle nonlinearity) and XGboost decoder which is SOTA for structured data regression and classification tasks. We study and contrast the methodologies with (variational) autoencoders in several toy low dimensional scenarios to derive necessary intuitions. The framework is applied to high dimensional simulated credit scoring data which parallels real-life financial applications. We applied the method to robustness testing to demonstrate practical use cases. The case study result suggests that the method provides an alternative to raw and quantile perturbation for model robustness testing. We show that the method is simplistic, guarantees interpretability all the way through, does not require extra tuning and provide unique benefits.

cs.LG↗

arXiv2023

Multi-resolution Interpretation and Diagnostics Tool for Natural Language Classifiers

Developing explainability methods for Natural Language Processing (NLP) models is a challenging task, for two main reasons. First, the high dimensionality of the data (large number of tokens) results in low coverage and in turn small contributions for the top tokens, compared to the overall model performance. Second, owing to their textual nature, the input variables, after appropriate transformations, are effectively binary (presence or absence of a token in an observation), making the input-output relationship difficult to understand. Common NLP interpretation techniques do not have flexibility in resolution, because they usually operate at word-level and provide fully local (message level) or fully global (over all messages) summaries. The goal of this paper is to create more flexible model explainability summaries by segments of observation or clusters of words that are semantically related to each other. In addition, we introduce a root cause analysis method for NLP models, by analyzing representative False Positive and False Negative examples from different segments. At the end, we illustrate, using a Yelp review data set with three segments (Restaurant, Hotel, and Beauty), that exploiting group/cluster structures in words and/or messages can aid in the interpretation of decisions made by NLP models and can be utilized to assess the model's sensitivity or bias towards gender, syntax, and word meanings.

cs.CL↗

arXiv2021

Model Explainability in Deep Learning Based Natural Language Processing

Machine learning (ML) model explainability has received growing attention, especially in the area related to model risk and regulations. In this paper, we reviewed and compared some popular ML model explainability methodologies, especially those related to Natural Language Processing (NLP) models. We then applied one of the NLP explainability methods Layer-wise Relevance Propagation (LRP) to a NLP classification model. We used the LRP method to derive a relevance score for each word in an instance, which is a local explainability. The relevance scores are then aggregated together to achieve global variable importance of the model. Through the case study, we also demonstrated how to apply the local explainability method to false positive and false negative instances to discover the weakness of a NLP model. These analysis can help us to understand NLP models better and reduce the risk due to the black-box nature of NLP models. We also identified some common issues due to the special natures of NLP models and discussed how explainability analysis can act as a control to detect these issues after the model has been trained.

cs.AI↗

arXiv2021

Bias, Fairness, and Accountability with AI and ML Algorithms

The advent of AI and ML algorithms has led to opportunities as well as challenges. In this paper, we provide an overview of bias and fairness issues that arise with the use of ML algorithms. We describe the types and sources of data bias, and discuss the nature of algorithmic unfairness. This is followed by a review of fairness metrics in the literature, discussion of their limitations, and a description of de-biasing (or mitigation) techniques in the model life cycle.

stat.ML↗

arXiv2021

Survey of Imbalanced Data Methodologies

Imbalanced data set is a problem often found and well-studied in financial industry. In this paper, we reviewed and compared some popular methodologies handling data imbalance. We then applied the under-sampling/over-sampling methodologies to several modeling algorithms on UCI and Keel data sets. The performance was analyzed for class-imbalance methods, modeling algorithms and grid search criteria comparison.

stat.ML↗

arXiv2020

Improving the Reconstruction of Disentangled Representation Learners via Multi-Stage Modeling

Current autoencoder-based disentangled representation learning methods achieve disentanglement by penalizing the (aggregate) posterior to encourage statistical independence of the latent factors. This approach introduces a trade-off between disentangled representation learning and reconstruction quality since the model does not have enough capacity to learn correlated latent variables that capture detail information present in most image data. To overcome this trade-off, we present a novel multi-stage modeling approach where the disentangled factors are first learned using a penalty-based disentangled representation learning method; then, the low-quality reconstruction is improved with another deep generative model that is trained to model the missing correlated latent variables, adding detail information while maintaining conditioning on the previously learned disentangled factors. Taken together, our multi-stage modelling approach results in a single, coherent probabilistic model that is theoretically justified by the principal of D-separation and can be realized with a variety of model classes including likelihood-based models such as variational autoencoders, implicit models such as generative adversarial networks, and tractable models like normalizing flows or mixtures of Gaussians. We demonstrate that our multi-stage model has higher reconstruction quality than current state-of-the-art methods with equivalent disentanglement performance across multiple standard benchmarks. In addition, we apply the multi-stage model to generate synthetic tabular datasets, showcasing an enhanced performance over benchmark models across a variety of metrics. The interpretability analysis further indicates that the multi-stage model can effectively uncover distinct and meaningful features of variations from which the original distribution can be recovered.

stat.ML↗

arXiv2010

Sparse regulatory networks

In many organisms the expression levels of each gene are controlled by the activation levels of known "Transcription Factors" (TF). A problem of considerable interest is that of estimating the "Transcription Regulation Networks" (TRN) relating the TFs and genes. While the expression levels of genes can be observed, the activation levels of the corresponding TFs are usually unknown, greatly increasing the difficulty of the problem. Based on previous experimental work, it is often the case that partial information about the TRN is available. For example, certain TFs may be known to regulate a given gene or in other cases a connection may be predicted with a certain probability. In general, the biology of the problem indicates there will be very few connections between TFs and genes. Several methods have been proposed for estimating TRNs. However, they all suffer from problems such as unrealistic assumptions about prior knowledge of the network structure or computational limitations. We propose a new approach that can directly utilize prior information about the network structure in conjunction with observed gene expression data to estimate the TRN. Our approach uses $L_1$ penalties on the network to ensure a sparse structure. This has the advantage of being computationally efficient as well as making many fewer assumptions about the network structure. We use our methodology to construct the TRN for E. coli and show that the estimate is biologically sensible and compares favorably with previous estimates.

stat.AP↗

arXiv2010

Group Variable Selection via a Hierarchical Lasso and Its Oracle Property

In many engineering and scientific applications, prediction variables are grouped, for example, in biological applications where assayed genes or proteins can be grouped by biological roles or biological pathways. Common statistical analysis methods such as ANOVA, factor analysis, and functional modeling with basis sets also exhibit natural variable groupings. Existing successful group variable selection methods such as Antoniadis and Fan (2001), Yuan and Lin (2006) and Zhao, Rocha and Yu (2009) have the limitation of selecting variables in an "all-in-all-out" fashion, i.e., when one variable in a group is selected, all other variables in the same group are also selected. In many real problems, however, we may want to keep the flexibility of selecting variables within a group, such as in gene-set selection. In this paper, we develop a new group variable selection method that not only removes unimportant groups effectively, but also keeps the flexibility of selecting variables within a group. We also show that the new method offers the potential for achieving the theoretical "oracle" property as in Fan and Li (2001) and Fan and Peng (2004).

stat.ME↗

arXiv2008

Partial Correlation Estimation by Joint Sparse Regression Models

In this paper, we propose a computationally efficient approach -- space(Sparse PArtial Correlation Estimation)-- for selecting non-zero partial correlations under the high-dimension-low-sample-size setting. This method assumes the overall sparsity of the partial correlation matrix and employs sparse regression techniques for model fitting. We illustrate the performance of space by extensive simulation studies. It is shown that space performs well in both non-zero partial correlation selection and the identification of hub variables, and also outperforms two existing methods. We then apply space to a microarray breast cancer data set and identify a set of hub genes which may provide important insights on genetic regulatory networks. Finally, we prove that, under a set of suitable assumptions, the proposed procedure is asymptotically consistent in terms of model selection and parameter estimation.

stat.ME↗