Search Results for author:
Found 25 papers, 6 papers with code
De-amplifying Bias from Differential Privacy in Language Model Fine-tuning
Through the case of binary gender bias, we demonstrate that Counterfactual Data Augmentation (CDA), a known method for addressing bias, also mitigates bias amplification by DP.
Is Certifying $\ell_p$ Robustness Still Worthwhile?
There are three layers to this inquiry, which we address in this paper: (1) why do we care about robustness research?
Identifying and Mitigating the Security Risks of Generative AI
However, GenAI can be used just as well by attackers to generate new attacks and increase the velocity and efficacy of existing attacks.
Order-sensitive Shapley Values for Evaluating Conceptual Soundness of NLP Models
Previous works show that deep NLP models are not always conceptually sound: they do not always learn the correct linguistic concepts.
Faithful Explanations for Deep Graph Models
Second, our analytical and empirical results demonstrate that feature attribution methods cannot capture the nonlinear effect of edge features, while existing subgraph explanation methods are not faithful.
Consistent Counterfactuals for Deep Models
Counterfactual examples are one of the most commonly-cited methods for explaining the predictions of machine learning models in key areas such as finance and medical diagnosis.
Robust Models Are More Interpretable Because Attributions Look Normal
Recent work has found that adversarially-robust deep networks used for image classification are more interpretable: their feature attributions tend to be sharper, and are more concentrated on the objects associated with the image's ground-truth class.
Influence Patterns for Explaining Information Flow in BERT
While attention is all you need may be proving true, we do not know why: attention-based transformer models such as BERT are superior but how information flows from input tokens to output predictions are unclear.
ABSTRACTING INFLUENCE PATHS FOR EXPLAINING (CONTEXTUALIZATION OF) BERT MODELS
While “attention is all you need” may be proving true, we do not yet know why: attention-based transformer models such as BERT are superior but how they contextualize information even for simple grammatical rules such as subject-verb number agreement(SVA) is uncertain.
Reconstructing Actions To Explain Deep Reinforcement Learning
Feature attribution has been a foundational building block for explaining the input feature importance in supervised learning with Deep Neural Network (DNNs), but face new challenges when applied to deep Reinforcement Learning (RL). We propose a new approach to explaining deep RL actions by defining a class of \emph{action reconstruction} functions that mimic the behavior of a network in deep RL.
Fairness Under Feature Exemptions: Counterfactual and Observational Measures
While quantifying disparity is essential, sometimes the needs of an occupation may require the use of certain features that are critical in a way that any disparity that can be explained by them might need to be exempted.
Smoothed Geometry for Robust Attribution
Feature attributions are a popular tool for explaining the behavior of Deep Neural Networks (DNNs), but have recently been shown to be vulnerable to attacks that produce divergent explanations for nearby inputs.
Influence Paths for Characterizing Subject-Verb Number Agreement in LSTM Language Models
LSTM-based recurrent neural networks are the state-of-the-art for many natural language processing (NLP) tasks.
Interpreting Interpretations: Organizing Attribution Methods by Criteria
In this work we expand the foundationsof human-understandable concepts with which attributionscan be interpreted beyond "importance" and its visualization; we incorporate the logical concepts of necessity andsufficiency, and the concept of proportionality.
Feature-Wise Bias Amplification
This overestimation gives rise to feature-wise bias amplification -- a previously unreported form of bias that can be traced back to the features of a trained model.
Hunting for Discriminatory Proxies in Linear Regression Models
In this paper we formulate a definition of proxy use for the setting of linear regression and present algorithms for detecting proxies.
Gender Bias in Neural Natural Language Processing
We define a general benchmark to quantify gender bias in a variety of neural NLP tasks.
Supervising Feature Influence
Standard methods for training classifiers that minimize empirical risk do not constrain the behavior of the classifier on such datapoints.
Influence-Directed Explanations for Deep Convolutional Networks
We study the problem of explaining a rich class of behavioral properties of deep neural networks.
Latent Factor Interpretations for Collaborative Filtering
The interpretation of latent factors can then replace the uninterpreted latent factors, resulting in a new model that expresses predictions in terms of interpretable features.
Case Study: Explaining Diabetic Retinopathy Detection Deep CNNs via Integrated Gradients
In this report, we applied integrated gradients to explaining a neural network for diabetic retinopathy detection.
Proxy Non-Discrimination in Data-Driven Systems
Machine learnt systems inherit biases against protected classes, historically disparaged groups, from training data.
Use Privacy in Data-Driven Systems: Theory and Experiments with Machine Learnt Programs
For a specific instantiation of this definition, we present a program analysis technique that detects instances of proxy use in a model, and provides a witness that identifies which parts of the corresponding program exhibit the behavior.
GOTCHA Password Hackers!
(2) The puzzles are hard for a computer to solve even if it has the random bits used by the computer to generate the final puzzle --- unlike a CAPTCHA.
Differentially Private Data Analysis of Social Networks via Restricted Sensitivity
Specifically, given a query f and a hypothesis H about the structure of a dataset D, we show generically how to transform f into a new query f_H whose global sensitivity (over all datasets including those that do not satisfy H) matches the restricted sensitivity of the query f. Moreover, if the belief of the querier is correct (i. e., D is in H) then f_H(D) = f(D).
Cryptography and Security Social and Information Networks Physics and Society
