Search Results for author:
Found 32 papers, 2 papers with code
Label Noise Robustness for Domain-Agnostic Fair Corrections via Nearest Neighbors Label Spreading
Last-layer retraining methods have emerged as an efficient framework for correcting existing base models.
Theoretical Guarantees of Data Augmented Last Layer Retraining Methods
Ensuring fair predictions across many distinct subpopulations in the training data can be prohibitive for large models.
Model Predictive Control for Joint Ramping and Regulation-Type Service from Distributed Energy Resource Aggregations
Distributed energy resources (DERs) such as grid-responsive loads and batteries can be harnessed to provide ramping and regulation services across the grid.
An Adversarial Approach to Evaluating the Robustness of Event Identification Models
Thorough experiments on the synthetic South Carolina 500-bus system highlight that a relatively simpler model such as logistic regression is more susceptible to adversarial attacks than gradient boosting.
Robustness to Subpopulation Shift with Domain Label Noise via Regularized Annotation of Domains
Existing methods for last layer retraining that aim to optimize worst-group accuracy (WGA) rely heavily on well-annotated groups in the training data.
Parameter Optimization with Conscious Allocation (POCA)
The performance of modern machine learning algorithms depends upon the selection of a set of hyperparameters.
Addressing GAN Training Instabilities via Tunable Classification Losses
Generalizing this dual-objective formulation using CPE losses, we define and obtain upper bounds on an appropriately defined estimation error.
A Semi-Supervised Approach for Power System Event Identification
Using this package, we generate and evaluate eventful PMU data for the South Carolina synthetic network.
$(α_D,α_G)$-GANs: Addressing GAN Training Instabilities via Dual Objectives
In an effort to address the training instabilities of GANs, we introduce a class of dual-objective GANs with different value functions (objectives) for the generator (G) and discriminator (D).
Smoothly Giving up: Robustness for Simple Models
There is a growing need for models that are interpretable and have reduced energy and computational cost (e. g., in health care analytics and federated learning).
Robust Model Selection of Gaussian Graphical Models
We first characterize the equivalence class up to which general graphs can be recovered in the presence of noise.
The Saddle-Point Accountant for Differential Privacy
SPA approximates privacy guarantees for the composition of DP mechanisms in an accurate and fast manner.
Parameter Estimation in Ill-conditioned Low-inertia Power Systems
This ill-conditioning is because of converter-interfaced power systems generators' zero or small inertia contribution.
Cactus Mechanisms: Optimal Differential Privacy Mechanisms in the Large-Composition Regime
Since the optimization problem is infinite dimensional, it cannot be solved directly; nevertheless, we quantize the problem to derive near-optimal additive mechanisms that we call "cactus mechanisms" due to their shape.
AugLoss: A Robust Augmentation-based Fine Tuning Methodology
As a step towards addressing both problems simultaneously, we introduce AugLoss, a simple but effective methodology that achieves robustness against both train-time noisy labeling and test-time feature distribution shifts by unifying data augmentation and robust loss functions.
$α$-GAN: Convergence and Estimation Guarantees
We prove a two-way correspondence between the min-max optimization of general CPE loss function GANs and the minimization of associated $f$-divergences.
A Machine Learning Framework for Event Identification via Modal Analysis of PMU Data
Using the obtained subset of features, we investigate the performance of two well-known classification models, namely, logistic regression (LR) and support vector machines (SVM) to identify generation loss and line trip events in two datasets.
Synthetic Time-Series Load Data via Conditional Generative Adversarial Networks
A framework for the generation of synthetic time-series transmission-level load data is presented.
Generation of Synthetic Multi-Resolution Time Series Load Data
The availability of large datasets is crucial for the development of new power system applications and tools; unfortunately, very few are publicly and freely available.
Being Properly Improper
Hence, optimizing a proper loss function on twisted data could perilously lead the learning algorithm towards the twisted posterior, rather than to the desired clean posterior.
Realizing GANs via a Tunable Loss Function
We introduce a tunable GAN, called $\alpha$-GAN, parameterized by $\alpha \in (0,\infty]$, which interpolates between various $f$-GANs and Integral Probability Metric based GANs (under constrained discriminator set).
A Verifiable Framework for Cyber-Physical Attacks and Countermeasures in a Resilient Electric Power Grid
In this paper, we investigate the feasibility and physical consequences of cyber attacks against energy management systems (EMS).
Three Variants of Differential Privacy: Lossless Conversion and Applications
In the first part, we develop a machinery for optimally relating approximate DP to RDP based on the joint range of two $f$-divergences that underlie the approximate DP and RDP.
On the alpha-loss Landscape in the Logistic Model
We analyze the optimization landscape of a recently introduced tunable class of loss functions called $\alpha$-loss, $\alpha \in (0,\infty]$, in the logistic model.
A Better Bound Gives a Hundred Rounds: Enhanced Privacy Guarantees via $f$-Divergences
We derive the optimal differential privacy (DP) parameters of a mechanism that satisfies a given level of R\'enyi differential privacy (RDP).
Theoretical Guarantees for Model Auditing with Finite Adversaries
Privacy concerns have led to the development of privacy-preserving approaches for learning models from sensitive data.
Generating Fair Universal Representations using Adversarial Models
We present a data-driven framework for learning fair universal representations (FUR) that guarantee statistical fairness for any learning task that may not be known a priori.
A Tunable Loss Function for Robust Classification: Calibration, Landscape, and Generalization
We introduce a tunable loss function called $\alpha$-loss, parameterized by $\alpha \in (0,\infty]$, which interpolates between the exponential loss ($\alpha = 1/2$), the log-loss ($\alpha = 1$), and the 0-1 loss ($\alpha = \infty$), for the machine learning setting of classification.
Generative Adversarial Models for Learning Private and Fair Representations
We present Generative Adversarial Privacy and Fairness (GAPF), a data-driven framework for learning private and fair representations of the data.
A Tunable Loss Function for Binary Classification
We present $\alpha$-loss, $\alpha \in [1,\infty]$, a tunable loss function for binary classification that bridges log-loss ($\alpha=1$) and $0$-$1$ loss ($\alpha = \infty$).
Generative Adversarial Privacy
We present a data-driven framework called generative adversarial privacy (GAP).
Context-Aware Generative Adversarial Privacy
On the one hand, context-free privacy solutions, such as differential privacy, provide strong privacy guarantees, but often lead to a significant reduction in utility.
