Shibal Ibrahim
YOU?
Author Swipe
View article: OSSCAR: One-Shot Structured Pruning in Vision and Language Models with Combinatorial Optimization
OSSCAR: One-Shot Structured Pruning in Vision and Language Models with Combinatorial Optimization Open
Structured pruning is a promising approach for reducing the inference costs of large vision and language models. By removing carefully chosen structures, e.g., neurons or attention heads, the improvements from this approach can be realized…
View article: Dyn-GWN: Time-Series Forecasting using Time-varying Graphs with Applications to Finance and Traffic Prediction
Dyn-GWN: Time-Series Forecasting using Time-varying Graphs with Applications to Finance and Traffic Prediction Open
Spatio-temporal modeling is an essential lens to understand many real-world phenomena from traffic to finance. There has been exciting work that explores spatio-temporal modeling with temporal graph convolutional networks. Often these meth…
View article: End-to-end Feature Selection Approach for Learning Skinny Trees
End-to-end Feature Selection Approach for Learning Skinny Trees Open
We propose a new optimization-based approach for feature selection in tree ensembles, an important problem in statistics and machine learning. Popular tree ensemble toolkits e.g., Gradient Boosted Trees and Random Forests support feature s…
View article: COMET: Learning Cardinality Constrained Mixture of Experts with Trees and Local Search
COMET: Learning Cardinality Constrained Mixture of Experts with Trees and Local Search Open
The sparse Mixture-of-Experts (Sparse-MoE) framework efficiently scales up model capacity in various domains, such as natural language processing and vision. Sparse-MoEs select a subset of the "experts" (thus, only a portion of the overall…
View article: Knowledge Graph Guided Simultaneous Forecasting and Network Learning for Multivariate Financial Time Series
Knowledge Graph Guided Simultaneous Forecasting and Network Learning for Multivariate Financial Time Series Open
Financial time series forecasting is challenging due to limited sample size, correlated samples, low signal strengths, among others. Additional information with knowledge graphs (KGs) can allow for improved prediction and decision making. …
View article: Flexible Modeling and Multitask Learning using Differentiable Tree Ensembles
Flexible Modeling and Multitask Learning using Differentiable Tree Ensembles Open
Decision tree ensembles are widely used and competitive learning models. Despite their success, popular toolkits for learning tree ensembles have limited modeling capabilities. For instance, these toolkits support a limited number of loss …
View article: Flexible Modeling and Multitask Learning using Differentiable Tree Ensembles
Flexible Modeling and Multitask Learning using Differentiable Tree Ensembles Open
Decision tree ensembles are widely used and competitive learning models. Despite their success, popular toolkits for learning tree ensembles have limited modeling capabilities. For instance, these toolkits support a limited number of loss …
View article: Newer is not always better: Rethinking transferability metrics, their peculiarities, stability and performance
Newer is not always better: Rethinking transferability metrics, their peculiarities, stability and performance Open
Fine-tuning of large pre-trained image and language models on small customized datasets has become increasingly popular for improved prediction and efficient use of limited resources. Fine-tuning requires identification of best models to t…
View article: Predicting Census Survey Response Rates via Interpretable Nonparametric Additive Models with Structured Interactions.
Predicting Census Survey Response Rates via Interpretable Nonparametric Additive Models with Structured Interactions. Open
Accurate and interpretable prediction of survey response rates is important from an operational standpoint. The US Census Bureau's well-known ROAM application uses principled statistical models trained on the US Census Planning Database da…
View article: Predicting Census Survey Response Rates With Parsimonious Additive Models and Structured Interactions
Predicting Census Survey Response Rates With Parsimonious Additive Models and Structured Interactions Open
In this paper, we consider the problem of predicting survey response rates using a family of flexible and interpretable nonparametric models. The study is motivated by the US Census Bureau's well-known ROAM application, which uses a linear…
View article: Nonparametric Finite Mixture Models with Possible Shape Constraints: A Cubic Newton Approach
Nonparametric Finite Mixture Models with Possible Shape Constraints: A Cubic Newton Approach Open
We explore computational aspects of maximum likelihood estimation of the mixture proportions of a nonparametric finite mixture model -- a convex optimization problem with old roots in statistics and a key member of the modern data analysis…