Classification with imbalance: A similarity-based method for predicting respiratory failure

Shrivastava, Harsh; Huddar, Vijay; Bhattacharya, Sakyajit; Rajan, Vaibhav

doi:10.1109/bibm.2015.7359773

Cited by 11 publications

(6 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The NGR architecture provides flexibility in handling such class imbalanced data. We can leverage the existing data augmentation techniques [7,9] or cost function balancing techniques between the different classes data points [33,4,5,38]. Besides, the sparsity induced in the neural network architecture because of the GcPn forces the NGRs to focus on important trends and thus provides a certain extent of robustness to noisy data.…”

Section: Conclusion Discussion and Future Workmentioning

confidence: 99%

“…For instance, the cases of infant death during the first year of life are rare compared to cases of surviving infants. Getting good performance on imbalanced data is a challenging problem and multiple techniques have been developed to assist existing learning algorithms [7,33,4]. In our case, we directly apply the models as is for obtaining results on the base implementation.…”

Section: Infant Mortality Data Analysismentioning

confidence: 99%

See 1 more Smart Citation

Neural Graph Revealers

Shrivastava¹,

Chajewska²

2023

Preprint

View full text Add to dashboard Cite

Sparse graph recovery methods work well where the data follows their assumptions but often they are not designed for doing downstream probabilistic queries. This limits their adoption to only identifying connections among the input variables. On the other hand, the Probabilistic Graphical Models (PGMs) assume an underlying base graph between variables and learns a distribution over them. PGM design choices are carefully made such that the inference & sampling algorithms are efficient. This brings in certain restrictions and often simplifying assumptions. In this work, we propose Neural Graph Revealers (NGRs), that are an attempt to efficiently merge the sparse graph recovery methods with PGMs into a single flow. The problem setting consists of an input data X with D features and M samples and the task is to recover a sparse graph showing connection between the features and learn a probability distribution over the D at the same time. NGRs view the neural networks as a 'glass box' or more specifically as a multitask learning framework. We introduce 'Graph-constrained path norm' that NGRs leverage to learn a graphical model that captures complex non-linear functional dependencies between the features in the form of an undirected sparse graph. Furthermore, NGRs can handle multimodal inputs like images, text, categorical data, embeddings etc. which is not straightforward to incorporate in the existing methods. We show experimental results of doing sparse graph recovery and probabilistic inference on data from Gaussian graphical models and a multimodal infant mortality dataset by Centers for Disease Control and Prevention.

show abstract

Section: Conclusion Discussion and Future Workmentioning

confidence: 99%

Section: Infant Mortality Data Analysismentioning

confidence: 99%

Neural Graph Revealers

Shrivastava¹,

Chajewska²

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…In this work, we investigate a new similarity‐based classification algorithm, called Q, for breast cancer diagnosis. The Q‐classifier is designed to learn from imbalanced datasets [35]. Unlike standard similarity‐based approaches, the Q‐classifier does not employ fixed parameters for the similarity function, and instead, it uses adaptive similarity parameters learnt in a manner that accounts for an imbalance in the data.…”

Section: Methodsmentioning

confidence: 99%

“…The notation of the j th dimension of the i th data point is

x_{i j}

, while the notation of the i th p‐dimensional data point is

x_{i}

. The Q‐classifier algorithm can be summarised in the following steps [35]: Landmark selection: select l landmark points from the training data by clustering the majority and minority class data points separately using the cluster centres as landmark points. The clustering operation is performed using the K‐means approach because it gives better results than other techniques such as SVM.…”

Section: Methodsmentioning

confidence: 99%

Fully automated scheme for computer‐aided detection and breast cancer diagnosis using digitised mammograms

Eltrass

Salama

2020

IET Image Processing

View full text Add to dashboard Cite

“…This will help identify key features and in-turn improve performance in cases where there is less data or imbal- anced data (more data points for one class over another). Some of the methods for class imbalance handling on which uGLAD model can act as a preprocessing steps are [35,40,4]. • Gaussian processes & time series problems: uGLAD can be extended to this interesting work by [7] on combining graphical lasso with Gaussian processes for learning gene regulatory networks.…”

Section: Potential Other Applications Of the Uglad Modelmentioning

confidence: 99%

uGLAD: Sparse graph recovery by optimizing deep unrolled networks

Shrivastava¹,

Chajewska²,

Abraham³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Probabilistic Graphical Models (PGMs) are generative models of complex systems. They rely on conditional independence assumptions between variables to learn sparse representations which can be visualized in a form of a graph. Such models are used for domain exploration and structure discovery in poorly understood domains. This work introduces a novel technique to perform sparse graph recovery by optimizing deep unrolled networks. Assuming that the input data X ∈ R M ×D comes from an underlying multivariate Gaussian distribution, we apply a deep model on X that outputs the precision matrix Θ, which can also be interpreted as the adjacency matrix. Our model, uGLAD 1 , builds upon and extends the state-ofthe-art model GLAD [42] to the unsupervised setting. The key benefits of our model are (1) uGLAD automatically optimizes sparsity-related regularization parameters leading to better performance than existing algorithms. (2) We introduce multitask learning based 'consensus' strategy for robust handling of missing data in an unsupervised setting. We evaluate model results on synthetic Gaussian data, non-Gaussian data generated from Gene Regulatory Networks, and present a case study in anaerobic digestion.

show abstract

Classification with imbalance: A similarity-based method for predicting respiratory failure

Cited by 11 publications

References 23 publications

Neural Graph Revealers

Neural Graph Revealers

Fully automated scheme for computer‐aided detection and breast cancer diagnosis using digitised mammograms

uGLAD: Sparse graph recovery by optimizing deep unrolled networks

Contact Info

Product

Resources

About