Demystifying local and global fairness trade-offs in federated learning using partial information decomposition
In this paper, we present an information-theoretic perspective to group fairness trade-offs in
federated learning (FL) with respect to sensitive attributes, such as gender, race, etc …
federated learning (FL) with respect to sensitive attributes, such as gender, race, etc …
Quantifying spuriousness of biased datasets using partial information decomposition
Spurious patterns refer to a mathematical association between two or more variables in a
dataset that are not causally related. However, this notion of spuriousness, which is usually …
dataset that are not causally related. However, this notion of spuriousness, which is usually …
Demystifying local and global fairness trade-offs in federated learning using information theory
We present an information-theoretic perspective to group fairness trade-offs in federated
learning (FL) with respect to sensitive attributes, such as gender, race, etc. Existing works …
learning (FL) with respect to sensitive attributes, such as gender, race, etc. Existing works …
Causal information splitting: Engineering proxy features for robustness to distribution shifts
B Mazaheri, A Mastakouri… - Uncertainty in …, 2023 - proceedings.mlr.press
Statistical prediction models are often trained on data that is drawn from different probability
distributions than their eventual use cases. One approach to proactively prepare for these …
distributions than their eventual use cases. One approach to proactively prepare for these …
A Unified View of Group Fairness Tradeoffs Using Partial Information Decomposition
This paper introduces a novel information-theoretic perspective on the relationship between
prominent group fairness notions in machine learning, namely statistical parity, equalized …
prominent group fairness notions in machine learning, namely statistical parity, equalized …
Quantifying Knowledge Distillation Using Partial Information Decomposition
Knowledge distillation provides an effective method for deploying complex machine learning
models in resource-constrained environments. It typically involves training a smaller student …
models in resource-constrained environments. It typically involves training a smaller student …
Combining Sources and Leveraging Contexts
B Mazaheri - 2024 - search.proquest.com
In this thesis we discuss two levels of knowledge beyond regression and classification. The
first involves the identification of exchangeable scenarios or individuals from which causal …
first involves the identification of exchangeable scenarios or individuals from which causal …
Analytically deriving Partial Information Decomposition for affine systems of stable and convolution-closed distributions
Bivariate partial information decomposition (PID) has emerged as a promising tool for
analyzing interactions in complex systems, particularly in neuroscience. PID achieves this by …
analyzing interactions in complex systems, particularly in neuroscience. PID achieves this by …
Gone With the Bits: Benchmarking Bias in Facial Phenotype Degradation Under Low-Rate Neural Compression
In this study, we investigate how facial phenotypes are distorted under neural image
compression and the disparity of this distortion across racial groups. Neural compression …
compression and the disparity of this distortion across racial groups. Neural compression …
Formalizing Limits of Knowledge Distillation Using Partial Information Decomposition
Knowledge distillation provides an effective method for deploying complex machine learning
models in resource-constrained environments. It typically involves training a smaller student …
models in resource-constrained environments. It typically involves training a smaller student …