Канал: Nicolas Papernot

SaTML 2024 - Javier Rando - Find the Trojan: Universal Backdoor Detection in Aligned LLMs

SaTML 2024 - Javier Rando - Find the Trojan: Universal Backdoor Detection in Aligned LLMs

SaTML 2024 - Stephen Casper - CNN Interpretability Competition

SaTML 2024 - Stephen Casper - CNN Interpretability Competition

SaTML 2024 - Edoardo Debenedetti & Daniel Paleka - Large Language Models Capture-the-Flag

SaTML 2024 - Edoardo Debenedetti & Daniel Paleka - Large Language Models Capture-the-Flag

SaTML 2024 - Carmela Troncoso & Nicolas Papernot - Closing Remarks

SaTML 2024 - Carmela Troncoso & Nicolas Papernot - Closing Remarks

SaTML 2024 - Chris Leckie - Shattering the Illusion of Unexploitable Data using Diffusion Models

SaTML 2024 - Chris Leckie - Shattering the Illusion of Unexploitable Data using Diffusion Models

SaTML 2024 - Eleanor Clifford - Backdoors in compiled neural networks

SaTML 2024 - Eleanor Clifford - Backdoors in compiled neural networks

SaTML 2024 - Yiwei Lu - Indiscriminate Data Poisoning Attacks on Pre-trained Feature Extractors

SaTML 2024 - Yiwei Lu - Indiscriminate Data Poisoning Attacks on Pre-trained Feature Extractors

SaTML 2024 - Hiroya Kato - EdgePruner: Poisoned Edge Pruning in Graph Contrastive Learning

SaTML 2024 - Hiroya Kato - EdgePruner: Poisoned Edge Pruning in Graph Contrastive Learning

SaTML 2024 - Quentin Le Roux - Black-Box Defense against DNN Backdoors with Rare Event Simulation

SaTML 2024 - Quentin Le Roux - Black-Box Defense against DNN Backdoors with Rare Event Simulation

SaTML 2024 - Ruinan Jin - Backdoor Attack on Un-paired Medical Image-Text Pretrained Models

SaTML 2024 - Ruinan Jin - Backdoor Attack on Un-paired Medical Image-Text Pretrained Models

SaTML 2024 - Francesco Croce - Prompt-Agnostic Evasion Attacks on Segmentation Models

SaTML 2024 - Francesco Croce - Prompt-Agnostic Evasion Attacks on Segmentation Models

SaTML 2024 - Sheila McIlraith - Tutorial - (Formal) Languages Help AI agents Learn and Reason

SaTML 2024 - Sheila McIlraith - Tutorial - (Formal) Languages Help AI agents Learn and Reason

SaTML 2024 - Edoardo Debenedetti - Evading Black-box Classifiers Without Breaking Eggs

SaTML 2024 - Edoardo Debenedetti - Evading Black-box Classifiers Without Breaking Eggs

SaTML 2024 - Anshuman Suri - SoK: Pitfalls in Evaluating Black-Box Attacks

SaTML 2024 - Anshuman Suri - SoK: Pitfalls in Evaluating Black-Box Attacks

SaTML 2024 - Chenxi Yang - Certifiably Robust RL through Model-Based Abstract Interpretation

SaTML 2024 - Chenxi Yang - Certifiably Robust RL through Model-Based Abstract Interpretation

SaTML 2024 - Ashutosh Kumar Nirala - Fast Certification Using Incremental Randomized Smoothing

SaTML 2024 - Ashutosh Kumar Nirala - Fast Certification Using Incremental Randomized Smoothing

SaTML 2024 - Lukas Fluri - Evaluating Superhuman Models with Consistency Checks

SaTML 2024 - Lukas Fluri - Evaluating Superhuman Models with Consistency Checks

SaTML 2024 - Kamala Varma - Understanding and Mitigating the Causes of Inference Slowdown for LMs

SaTML 2024 - Kamala Varma - Understanding and Mitigating the Causes of Inference Slowdown for LMs

SaTML 2024 - Nishtha Madaan - Counterfactual Generation via Structured Diffusion

SaTML 2024 - Nishtha Madaan - Counterfactual Generation via Structured Diffusion

SaTML 2024 - Lea Schönherr - CodeLMSec: Finding Security Vulnerabilities in Black-Box Code LMs

SaTML 2024 - Lea Schönherr - CodeLMSec: Finding Security Vulnerabilities in Black-Box Code LMs

SaTML 2024 - Judy Hanwen Shen - SoK: Unifying Corroborative and Contributive Attributions in LLMs

SaTML 2024 - Judy Hanwen Shen - SoK: Unifying Corroborative and Contributive Attributions in LLMs

SaTML 2024 - Yves-Alexandre de Montjoye - Tutorial - Detecting the use of copyright content by LLMs

SaTML 2024 - Yves-Alexandre de Montjoye - Tutorial - Detecting the use of copyright content by LLMs

SaTML 2024 - Patrick Vossler - Conventional Fairness Metrics With Probabilistic Protected Features

SaTML 2024 - Patrick Vossler - Conventional Fairness Metrics With Probabilistic Protected Features

SaTML 2024 - Shengyuan Hu - Fair Federated Learning via Bounded Group Loss

SaTML 2024 - Shengyuan Hu - Fair Federated Learning via Bounded Group Loss

SaTML 2024 - Wenxin Ding - Towards Scalable and Robust Model Versioning

SaTML 2024 - Wenxin Ding - Towards Scalable and Robust Model Versioning

SaTML 2024 - Zhifeng Kong - Data Redaction from Conditional Generative Models

SaTML 2024 - Zhifeng Kong - Data Redaction from Conditional Generative Models

SaTML 2024 - Andrew Geng - Model Reprogramming Outperforms Fine-tuning on OOD in Text-Image Encoders

SaTML 2024 - Andrew Geng - Model Reprogramming Outperforms Fine-tuning on OOD in Text-Image Encoders

SaTML 2024 - Deborah Raji - Keynote - Audits and Accountability in the Age of 'AI'

SaTML 2024 - Deborah Raji - Keynote - Audits and Accountability in the Age of 'AI'

SaTML 2024 - Augustin Godinot - Under manipulations, are there AI models harder to audit?

SaTML 2024 - Augustin Godinot - Under manipulations, are there AI models harder to audit?

SaTML 2024 - Ryan Steed - SoK: AI Auditing: The Broken Bus on the Road to AI Accountability

SaTML 2024 - Ryan Steed - SoK: AI Auditing: The Broken Bus on the Road to AI Accountability

SaTML 2024 - Joseph Near - OLYMPIA: A Simulation Framework for Evaluating Secure Aggregation

SaTML 2024 - Joseph Near - OLYMPIA: A Simulation Framework for Evaluating Secure Aggregation

SaTML 2024- Audra McMillan - Differentially Private Heavy Hitter Detection using Federated Analytics

SaTML 2024- Audra McMillan - Differentially Private Heavy Hitter Detection using Federated Analytics

SaTML 2024 - Ajith Suresh - ScionFL: Efficient and Robust Secure Quantized Aggregation

SaTML 2024 - Ajith Suresh - ScionFL: Efficient and Robust Secure Quantized Aggregation

SaTML 2024 - Tatsuki Koga - Differentially Private Multi-Site Treatment Effect Estimation

SaTML 2024 - Tatsuki Koga - Differentially Private Multi-Site Treatment Effect Estimation

SaTML 2024 - Chulin Xie - Improving Private-Vertical FL by Efficient Communication with ADMM

SaTML 2024 - Chulin Xie - Improving Private-Vertical FL by Efficient Communication with ADMM

SaTML 2024 - Francesco Pinto - PILLAR: How to make semi-private learning more effective

SaTML 2024 - Francesco Pinto - PILLAR: How to make semi-private learning more effective

SaTML 2024 - Achraf Azize - Concentrated Differential Privacy for Bandits

SaTML 2024 - Achraf Azize - Concentrated Differential Privacy for Bandits

SaTML 2024 - Amol Khanna - SoK: Differentially Private Linear Models For High Dimensional Data

SaTML 2024 - Amol Khanna - SoK: Differentially Private Linear Models For High Dimensional Data

SaTML 2024 - Shuai Tang - Improved Differentially Private Regression via Gradient Boosting

SaTML 2024 - Shuai Tang - Improved Differentially Private Regression via Gradient Boosting

SaTML 2024 - Zhangheng Li - Shake to Leak: Amplifying the Generative Privacy Risk through Finetuning

SaTML 2024 - Zhangheng Li - Shake to Leak: Amplifying the Generative Privacy Risk through Finetuning

SaTML 2024 - Julien Ferry - Probabilistic Dataset Reconstruction from Interpretable Models

SaTML 2024 - Julien Ferry - Probabilistic Dataset Reconstruction from Interpretable Models

SaTML 2024 - Somesh Jha - Keynote - Watermarking (The State of the Union)

SaTML 2024 - Somesh Jha - Keynote - Watermarking (The State of the Union)

SaTML 2024 - Carmela Troncoso and Nicolas Papernot - Opening Remarks

SaTML 2024 - Carmela Troncoso and Nicolas Papernot - Opening Remarks

SaTML 2023 - Edoardo Debenedetti - A Light Recipe to Train Robust Vision Transformers

SaTML 2023 - Edoardo Debenedetti - A Light Recipe to Train Robust Vision Transformers

SaTML 2023 - Giovanni Apruzzese - Real Attackers Don’t Compute Gradients

SaTML 2023 - Giovanni Apruzzese - Real Attackers Don’t Compute Gradients

SaTML 2023 - Zico Kolter - Robustness in Machine Learning: A Five-Year Retrospective

SaTML 2023 - Zico Kolter - Robustness in Machine Learning: A Five-Year Retrospective

SaTML 2023 - Xiaojun Xu - EDoG: Adversarial Edge Detection For Graph Neural Networks (virtual)

SaTML 2023 - Xiaojun Xu - EDoG: Adversarial Edge Detection For Graph Neural Networks (virtual)

SaTML 2023 - Sanghyun Hong - Publishing Efficient On-device Models Increases Adversarial Vuln.

SaTML 2023 - Sanghyun Hong - Publishing Efficient On-device Models Increases Adversarial Vuln.

SaTML 2023 - Katharina Beckh - SoK: Harnessing Prior Knowledge for Explainable Machine Learning

SaTML 2023 - Katharina Beckh - SoK: Harnessing Prior Knowledge for Explainable Machine Learning

SaTML 2023 - Jacob Steinhardt - Aligning ML Systems with Human Intent

SaTML 2023 - Jacob Steinhardt - Aligning ML Systems with Human Intent

SaTML 2023 - Washington Garcia - Less is More: Dimension Reduction Finds On-Manifold Adversarial Ex

SaTML 2023 - Washington Garcia - Less is More: Dimension Reduction Finds On-Manifold Adversarial Ex

SaTML 2023 - J. Zhang - CARE: Certifiably Robust Learning with Reasoning via Variational Inference

SaTML 2023 - J. Zhang - CARE: Certifiably Robust Learning with Reasoning via Variational Inference

SaTML 2023 - Reza Nasirigerdeh - Kernel Normalized Convolutional Networks for Privacy-Preserving ML

SaTML 2023 - Reza Nasirigerdeh - Kernel Normalized Convolutional Networks for Privacy-Preserving ML

SaTML 2023 - Ari Karchmer - Theoretical Limits of Provable Security Against Model Extraction

SaTML 2023 - Ari Karchmer - Theoretical Limits of Provable Security Against Model Extraction

SaTML 2023 - Zayd Hammoudeh - Reducing Certified Regression to Certified Classification

SaTML 2023 - Zayd Hammoudeh - Reducing Certified Regression to Certified Classification

SaTML 2023 - Julien Ferry - Exploiting Fairness to Enhance Sensitive Attributes Reconstruction

SaTML 2023 - Julien Ferry - Exploiting Fairness to Enhance Sensitive Attributes Reconstruction

SaTML 2023 - Timnit Gebru - Eugenics and the Promise of Utopia through AGI

SaTML 2023 - Timnit Gebru - Eugenics and the Promise of Utopia through AGI

SaTML 2023 - Yao Qin - What Are Effective Labels for Augmented Data?

SaTML 2023 - Yao Qin - What Are Effective Labels for Augmented Data?

SaTML 2023 - Lorenzo Cazzaro - Explainable Global Fairness Verification of Tree-Based Classifiers

SaTML 2023 - Lorenzo Cazzaro - Explainable Global Fairness Verification of Tree-Based Classifiers

SaTML 2023 - Teresa Datta - Position: Tensions Between the Proxies of Human Values in AI

SaTML 2023 - Teresa Datta - Position: Tensions Between the Proxies of Human Values in AI

SaTML 2023 - Arif - Reprogrammable-FL: Improving Utility-Priv Tradeoff in FL via Model Reprogramming

SaTML 2023 - Arif - Reprogrammable-FL: Improving Utility-Priv Tradeoff in FL via Model Reprogramming

SaTML 2023 - Keith Manville and Deepesh Chaudhari - Model Attribution Challenge

SaTML 2023 - Keith Manville and Deepesh Chaudhari - Model Attribution Challenge

SaTML 2023 - Nandakumar - PolyKervNets: Activation-free Neural Networks For Private Interference

SaTML 2023 - Nandakumar - PolyKervNets: Activation-free Neural Networks For Private Interference

SaTML 2023 - Liu - Towards Transferable Unrestricted Adversarial Examples with Minimum Changes

SaTML 2023 - Liu - Towards Transferable Unrestricted Adversarial Examples with Minimum Changes

SaTML 2023 - Carlini & Jagielski - Improving train data extraction attacks on large language models

SaTML 2023 - Carlini & Jagielski - Improving train data extraction attacks on large language models

SaTML 2023 - Minseon Kim - Rethinking the Entropy of Instance in Adversarial Training

SaTML 2023 - Minseon Kim - Rethinking the Entropy of Instance in Adversarial Training

SaTML 2023 - Nicolas Papernot - Closing Remarks

SaTML 2023 - Nicolas Papernot - Closing Remarks

SaTML 2023 - Yingyan Zeng - ModelPred: A Framework for Predicting Trained Model from Training Data

SaTML 2023 - Yingyan Zeng - ModelPred: A Framework for Predicting Trained Model from Training Data

SaTML 2023 - Kenfack - Learning Fair Representations thr. Uniformly Distributed Sensitive Attributes

SaTML 2023 - Kenfack - Learning Fair Representations thr. Uniformly Distributed Sensitive Attributes

SaTML 2023 - Guy Heller - Can Stochastic Grad Langevin Dynamics Provide Differential Privacy for DL?

SaTML 2023 - Guy Heller - Can Stochastic Grad Langevin Dynamics Provide Differential Privacy for DL?

SaTML 2023 - Gautam Kamath - An Introduction to Differential Privacy

SaTML 2023 - Gautam Kamath - An Introduction to Differential Privacy

SaTML 2023 - Kamalika Chaudhuri - Data Redaction from Pre-trained GANs

SaTML 2023 - Kamalika Chaudhuri - Data Redaction from Pre-trained GANs

SaTML 2023 - Krishna Acharya - Wealth Dynamics Over Generations: Analysis and Interventions

SaTML 2023 - Krishna Acharya - Wealth Dynamics Over Generations: Analysis and Interventions

SaTML 2023 - Florian Jaeckle - Neural Lower Bounds For Verification

SaTML 2023 - Florian Jaeckle - Neural Lower Bounds For Verification

SaTML 2023 - Anshuman Suri - Dissecting Distribution Inference

SaTML 2023 - Anshuman Suri - Dissecting Distribution Inference

SaTML 2023 - Mintong Kang - FaShapley Based Model Pruning Towards Certifiably Robust DNNs

SaTML 2023 - Mintong Kang - FaShapley Based Model Pruning Towards Certifiably Robust DNNs

SaTML 2023 - Sanjay Kariyappa - ExPLoit: Extracting Private Labels in Split Learning

SaTML 2023 - Sanjay Kariyappa - ExPLoit: Extracting Private Labels in Split Learning

SaTML 2023 - Harsh Chaudhari - SafeNet: The Unreasonable Effectiveness of Ensembles in Private CL

SaTML 2023 - Harsh Chaudhari - SafeNet: The Unreasonable Effectiveness of Ensembles in Private CL

SaTML 2023 - Kawakami - Validity Perspec on Evaluating Justified Use of Data-driven Decision-making

SaTML 2023 - Kawakami - Validity Perspec on Evaluating Justified Use of Data-driven Decision-making

SaTML 2023 - Patrick Altmeyer - Endogenous Macrodynamics in Algorithic Recourse

SaTML 2023 - Patrick Altmeyer - Endogenous Macrodynamics in Algorithic Recourse

SaTML 2023 - Korbinian Koch - Machine Unlearning with SISA Comes at the Expense of Minority Classes

SaTML 2023 - Korbinian Koch - Machine Unlearning with SISA Comes at the Expense of Minority Classes

SaTML 2023 - Stephen Casper - SoK: Toward Transparent AI: Interpreting the Inner Structures of DNNs

SaTML 2023 - Stephen Casper - SoK: Toward Transparent AI: Interpreting the Inner Structures of DNNs

SaTML 2023 - Sayanton Dibbo - Model Inversion Attack with Least Information

SaTML 2023 - Sayanton Dibbo - Model Inversion Attack with Least Information

SaTML 2023 - Nicolas Papernot - Opening Remarks

SaTML 2023 - Nicolas Papernot - Opening Remarks

SaTML 2023 - Hartmann - Distribution inference risks: Identifying and mitigating sources of leakage

SaTML 2023 - Hartmann - Distribution inference risks: Identifying and mitigating sources of leakage

SaTML 2023 - James Bailey - Backdoor Attacks on Time Series: A Generative Approach

SaTML 2023 - James Bailey - Backdoor Attacks on Time Series: A Generative Approach

SaTML 2023 - Hojjat Aghakhani - VENOMAVE: Targeted Poisoning Against Speech Recognition

SaTML 2023 - Hojjat Aghakhani - VENOMAVE: Targeted Poisoning Against Speech Recognition

SaTML 2023 - Juba Ziani - Optimal Data Acquisition with Privacy-Aware Agents

SaTML 2023 - Juba Ziani - Optimal Data Acquisition with Privacy-Aware Agents

SaTML 2023 - Gorka Abad - Sniper Backdoor: Single Client Targeted Backdoor Attack in FL

SaTML 2023 - Gorka Abad - Sniper Backdoor: Single Client Targeted Backdoor Attack in FL

SaTML 2023 - Wu and Tagomori - Toward Certified Robustness Against Real-World Distribution Shift

SaTML 2023 - Wu and Tagomori - Toward Certified Robustness Against Real-World Distribution Shift

The Role of Randomization in Trustworthy Machine Learning

The Role of Randomization in Trustworthy Machine Learning

1st IEEE Conference on Secure and Trustworthy Machine Learning (SaTML 2023)

1st IEEE Conference on Secure and Trustworthy Machine Learning (SaTML 2023)

What does it mean for ML to be trustworthy (June 2021 update)

What does it mean for ML to be trustworthy (June 2021 update)

Tempered Sigmoid Activations for Deep Learning with Differential Privacy

Tempered Sigmoid Activations for Deep Learning with Differential Privacy

What does it mean for ML to be trustworthy?

What does it mean for ML to be trustworthy?

How Relevant is the Turing Test in the Age of Sophisbots?

How Relevant is the Turing Test in the Age of Sophisbots?

SecML18: Somesh Jha on Semantic Adversarial Examples

SecML18: Somesh Jha on Semantic Adversarial Examples

SecML18: Been Kim on Interpretability for when NOT to use machine learning

SecML18: Been Kim on Interpretability for when NOT to use machine learning

SecML18: Csaba Szepesvari on Rigorous Agent Evaluation

SecML18: Csaba Szepesvari on Rigorous Agent Evaluation

SecML18: Marta Kwiatkowska on Safety verification for neural networks with provable guarantees

SecML18: Marta Kwiatkowska on Safety verification for neural networks with provable guarantees

SecML18: Arjun Nitin Bhagoji on Model Poisoning Attacks in Federated Learning

SecML18: Arjun Nitin Bhagoji on Model Poisoning Attacks in Federated Learning

SecML18: danah boyd on A Sociotechnical Approach to Security in Machine Learning

SecML18: danah boyd on A Sociotechnical Approach to Security in Machine Learning

SecML18: Sven Gowal on Interval Bound Propagation for Training Verifiably Robust Models

SecML18: Sven Gowal on Interval Bound Propagation for Training Verifiably Robust Models

SecML18: Aditi Raghunathan on Semidefinite relaxations for certifying robustness

SecML18: Aditi Raghunathan on Semidefinite relaxations for certifying robustness

SecML18: Salome Viljoen on Law and Adversarial Machine Learning

SecML18: Salome Viljoen on Law and Adversarial Machine Learning

SecML18: Jerry Li on Sever: A Robust Meta-Algorithm for Stochastic Optimization

SecML18: Jerry Li on Sever: A Robust Meta-Algorithm for Stochastic Optimization

ICLR oral - Semi-supervised Knowledge Transfer for Deep Learning from Private Training Data

ICLR oral - Semi-supervised Knowledge Transfer for Deep Learning from Private Training Data