1 citations · 3 across the 10 of their papers we have counts for
10 papers
Understanding Generative AI Content with Embedding Models
Max Vargas, Reilly Cannon, Andrew Engel +2
Constructing high-quality features is critical to any quantitative data analysis. While feature engineering was historically addressed by carefully hand-crafting data representatio…
Measuring training variability from stochastic optimization using robust nonparametric testing
Sinjini Banerjee, Tim Marrinan, Reilly Cannon +2
Deep neural network training often involves stochastic optimization, meaning each run will produce a different model. This implies that hyperparameters of the training process, suc…
Efficient kernel surrogates for neural network-based regression
Saad Qadeer, Andrew Engel, Amanda Howard +4
Despite their immense promise in performing a variety of learning tasks, a theoretical understanding of the limitations of Deep Neural Networks (DNNs) has so far eluded practitione…
Foundation Model's Embedded Representations May Detect Distribution Shift
Max Vargas, Adam Tsou, Andrew Engel +1
Sampling biases can cause distribution shifts between train and test datasets for supervised learning tasks, obscuring our ability to understand the generalization capacity of a mo…
Robust Nonparametric Hypothesis Testing to Understand Variability in Training Neural Networks
Sinjini Banerjee, Reilly Cannon, Tim Marrinan +2
Training a deep neural network (DNN) often involves stochastic optimization, which means each run will produce a different model. Several works suggest this variability is negligib…
Exploring Learned Representations of Neural Networks with Principal Component Analysis
Amit Harlev, Andrew Engel, Panos Stinis +1
Understanding feature representation for deep neural networks (DNNs) remains an open question within the general field of explainable AI. We use principal component analysis (PCA)…
Minibatching Offers Improved Generalization Performance for Second Order Optimizers
Eric Silk, Swarnita Chakraborty, Nairanjana Dasgupta +3
Training deep neural networks (DNNs) used in modern machine learning is computationally expensive. Machine learning scientists, therefore, rely on stochastic first-order methods fo…
Faithful and Efficient Explanations for Neural Networks via Neural Tangent Kernel Surrogate Models
Andrew Engel, Zhichao Wang, Natalie S. Frank +4
A recent trend in explainable AI research has focused on surrogate modeling, where neural networks are approximated as simpler ML algorithms such as kernel machines. A second trend…
Spectral Evolution and Invariance in Linear-width Neural Networks
Zhichao Wang, Andrew Engel, Anand Sarwate +2
We investigate the spectral properties of linear-width feed-forward neural networks, where the sample size is asymptotically proportional to network width. Empirically, we show tha…
TorchNTK: A Library for Calculation of Neural Tangent Kernels of PyTorch Models
Andrew Engel, Zhichao Wang, Anand D. Sarwate +2
We introduce torchNTK, a python library to calculate the empirical neural tangent kernel (NTK) of neural network models in the PyTorch framework. We provide an efficient method to…