Search Results for author:
Found 35 papers, 9 papers with code
Robustness and Equivariance of Neural Networks
Neural networks models are known to be vulnerable to geometric transformations as well as small pixel-wise perturbations of input.
NICE: To Optimize In-Context Examples or Not?
Recent work shows that in-context learning and optimization of in-context examples (ICE) can significantly improve the accuracy of large language models (LLMs) on a wide range of tasks, leading to an apparent consensus that ICE optimization is crucial for better performance.
Rethinking Robustness of Model Attributions
We observe two main causes for fragile attributions: first, the existing metrics of robustness (e. g., top-k intersection) over-penalize even reasonable local shifts in attribution, thereby making random perturbations to appear as a strong attack, and second, the attribution can be concentrated in a small region even when there are multiple important parts in an image.
How Far Can Fairness Constraints Help Recover From Biased Data?
We further generalize it to arbitrary data distributions and arbitrary hypothesis classes, i. e., we prove that for any data distribution, if the optimally accurate classifier in a given hypothesis class is fair and robust, then it can be recovered through fair classification with equal opportunity constraints on the biased distribution whenever the bias parameters satisfy certain simple conditions.
Improved Outlier Robust Seeding for k-means
However, in the presence of adversarial noise or outliers, $D^{2}$ sampling is more likely to pick centers from distant outliers instead of inlier clusters, and therefore its approximation guarantees \textit{w. r. t.}
Optimizing Group-Fair Plackett-Luce Ranking Models for Relevance and Ex-Post Fairness
Previous works have proposed efficient algorithms to train stochastic ranking models that achieve fairness of exposure to the groups ex-ante (or, in expectation), which may not guarantee representation fairness to the groups ex-post, that is, after realizing a ranking from the stochastic ranking model.
Sampling Individually-Fair Rankings that are Always Group Fair
Fair ranking tasks, which ask to rank a set of items to maximize utility subject to satisfying group-fairness constraints, have gained significant interest in the Algorithmic Fairness, Information Retrieval, and Machine Learning literature.
Causal Effect Regularization: Automated Detection and Removal of Spurious Attributes
We prove that our method only requires that the ranking of estimated causal effects is correct across attributes to select the correct classifier.
On Comparing Fair Classifiers under Data Bias
In this paper, we consider a theoretical model for injecting data bias, namely, under-representation and label bias (Blum & Stangl, 2019).
Identifying, measuring, and mitigating individual unfairness for supervised learning models and application to credit risk models
Furthermore, we developed a second metric (distinct from the fair similarity metric) to determine how fairly a model is treating similar individuals.
Socially Fair Center-based and Linear Subspace Clustering
Center-based clustering (e. g., $k$-means, $k$-medians) and clustering using linear subspaces are two most popular techniques to partition real-world data into smaller clusters.
One-pass additive-error subset selection for $\ell_{p}$ subspace approximation
In this paper, we give a one-pass subset selection with an additive approximation guarantee for $\ell_{p}$ subspace approximation, for any $p \in [1, \infty)$.
Sampling Ex-Post Group-Fair Rankings
Our second random walk-based algorithm samples ex-post group-fair rankings from a distribution $\delta$-close to $D$ in total variation distance and has expected running time $O^*(k^2\ell^2)$, when there is a sufficient gap between the given upper and lower bounds on the group-wise representation.
Learning and Generalization in Overparameterized Normalizing Flows
In supervised learning, it is known that overparameterized neural networks with one hidden layer provably and efficiently learn and generalize, when trained using stochastic gradient descent with a sufficiently small learning rate and suitable initialization.
Rawlsian Fair Adaptation of Deep Learning Classifiers
Given any score function or feature representation and only its second-order statistics on the sensitive sub-populations, we seek a threshold classifier on the given score or a linear threshold classifier on the given feature representation that achieves the Rawls error rate restricted to this hypothesis class.
On Subspace Approximation and Subset Selection in Fewer Passes by MCMC Sampling
Our ideas also extend to give a reduction in the number of passes required by adaptive sampling algorithms for $\ell_{p}$ subspace approximation and subset selection, for $p \geq 2$.
Learning and Generalization in Univariate Overparameterized Normalizing Flows
In supervised learning, it is known that overparameterized neural networks with one hidden layer provably and efficiently learn and generalize, when trained using Stochastic Gradient Descent (SGD).
The Importance of Modeling Data Missingness in Algorithmic Fairness: A Causal Perspective
For example, in multi-stage settings where decisions are made in multiple screening rounds, we use our framework to derive the minimal distributions required to design a fair algorithm.
On the Problem of Underranking in Group-Fair Ranking
We give a fair ranking algorithm that takes any given ranking and outputs another ranking with simultaneous underranking and group fairness guarantees comparable to the lower bound we prove.
Subspace approximation with outliers
Any multiplicative approximation algorithm for the subspace approximation problem with outliers must solve the robust subspace recovery problem, a special case in which the $(1-\alpha)n$ inliers in the optimal solution are promised to lie exactly on a $k$-dimensional linear subspace.
How do SGD hyperparameters in natural training affect adversarial robustness?
We observe that networks trained with constant learning rate to batch size ratio, as proposed in Jastrzebski et al., yield models which generalize well and also have almost constant adversarial robustness, independent of the batch size.
On Universalized Adversarial and Invariant Perturbations
Recent work by authors arXiv:2002. 11318 studies a trade-off between invariance and robustness to adversarial attacks.
Universalization of any adversarial attack using very few test examples
For VGG16 and VGG19 models trained on ImageNet, our simple universalization of Gradient, FGSM, and DeepFool perturbations using a test sample of 64 images gives fooling rates comparable to state-of-the-art universal attacks \cite{Dezfooli17, Khrulkov18} for reasonable norms of perturbation.
Can we have it all? On the Trade-off between Spatial and Adversarial Robustness of Neural Networks
(Non-)robustness of neural networks to small, adversarial pixel-wise perturbations, and as more recently shown, to even random spatial transformations (e. g., translations, rotations) entreats both theoretical and empirical understanding.
Invariance vs Robustness of Neural Networks
We observe that the rotation invariance of equivariant models (StdCNNs and GCNNs) improves by training augmentation with progressively larger rotations but while doing so, their adversarial robustness does not improve, or worse, it can even drop significantly on datasets such as MNIST.
Universal Adversarial Attack Using Very Few Test Examples
We evaluate the error rates and fooling rates of three universal attacks, SVD-Gradient, SVD-DeepFool and SVD-FGSM, on state of the art neural networks.
Quantifying Infra-Marginality and Its Trade-off with Group Fairness
To quantify this bias, we propose a general notion of $\eta$-infra-marginality that can be used to evaluate the extent of this bias.
Universal Attacks on Equivariant Networks
In this paper, we observe an interesting spectral property shared by all of the above input-dependent, pixel-wise adversarial attacks on translation and rotation-equivariant networks.
On Euclidean $k$-Means Clustering with $α$-Center Proximity
On the hardness side we show that for any $\alpha' > 1$, there exists an $\alpha \leq \alpha'$, $(\alpha >1)$, and an $\varepsilon_0 > 0$ such that minimizing the $k$-means objective over clusterings that satisfy $\alpha$-center proximity is NP-hard to approximate within a multiplicative $(1+\varepsilon_0)$ factor.
Fair and Diverse DPP-based Data Summarization
Sampling methods that choose a subset of the data proportional to its diversity in the feature space are popular for data summarization.
Depth separation and weight-width trade-offs for sigmoidal neural networks
We show a similar separation between the expressive power of depth-2 and depth-3 sigmoidal neural networks over a large class of input distributions, as long as the weights are polynomially bounded.
Batched Gaussian Process Bandit Optimization via Determinantal Point Processes
Gaussian Process bandit optimization has emerged as a powerful tool for optimizing noisy black box functions.
How to be Fair and Diverse?
However, in doing so, a question that seems to be overlooked is whether it is possible to produce fair subsamples that are also adequately representative of the feature space of the data set - an important and classic requirement in machine learning.
On the Complexity of Constrained Determinantal Point Processes
Consequently, we obtain a few algorithms of independent interest: 1) to count over the base polytope of regular matroids when there are additional (succinct) budget constraints and, 2) to evaluate and compute the mixed characteristic polynomials, that played a central role in the resolution of the Kadison-Singer problem, for certain special cases.
On Sampling and Greedy MAP Inference of Constrained Determinantal Point Processes
When pairwise similarities are captured by a kernel, the determinants of submatrices provide a measure of diversity or independence of items within a subset.
