Piotr (Peter) Mardziel
Piotr (Peter) Mardziel
About me
Independent researcher and software engineer. My research is mostly inactive except for reviewing in choice areas: transparency and accountability in machine learning with applications to security, privacy, and fairness; probabilistic programming and quantitative information flow.NOTE TO EDITORS: I will review articles only in open access journals hosted by non-profit organizations.
Unsorted Works
- Technical Exploration: How much pi does gpt know?
- Technical Blog Article: Drift Metrics: How to Select the Right Metric to Analyze Drift
- Technical Blog Article: Drift in Machine Learning: Why it's hard and what to do about it
- Software: trulens. A python library for explaining deep models (trulens_explain) and evaluating large language model applications (trulens_eval).
Software Engineering
I'm presently software engineering mostly on trulens, an open-source tool for evaluating generative models.Research
papers.txt; papers.html; google scholar; dblp; conflicts.txt; bio.txtS&P
Topics:
- Anonymity
- Language-based security
- Machine learning and AI security
- Privacy technologies and mechanisms
- Secure information flow
CSF
Topics:
- accountability
- anonymity and privacy
- formal methods and verification
- information flow control
- language-based security
- machine learning
- software security
CMU
Topics:
- Security and Privacy: Foundations and Formal Methods
- Application Domains: Machine learning
- Data Science: Machine Learning
- Data Science: Deep Learning
ML Security, Privacy, Fairness
- De-amplifying Bias from Differential Privacy in Language Model Fine-tuning. In PPAI 2024.
- Gender Bias in Neural Natural Language Processing. In Logic, Language, and Security: Essays Dedicated to Andre Scedrov on the Occasion of His 65th Birthday 2020.
- Smoothed Geometry for Robust Attribution. In NeurIPS 2020.
- An Information-Theoretic Quantification of Discrimination with Exempt Features. In AAAI 2020.
- Use Privacy in Data-Driven Systems: Theory and Experiments with Machine Learnt Programs. In CCS 2017.
ML Interpretability
- Influence Paths for Characterizing Subject-Verb Number Agreement in LSTM Language Models. In ACL 2020.
- Interpreting Interpretations: Organizing Attribution Methods by Criteria. In CVPR Workshop on Fair, Data Efficient and Trusted Computer Vision 2020.
- Score-CAM: Score-Weighted Visual Explanations for Convolutional Neural Networks. In CVPR Workshop on Fair, Data Efficient and Trusted Computer Vision 2020.
Security and Privacy
- Probabilistic Abstract Interpretation: Sound Inference and Application to Privacy. In Foundations of Probabilistic Programming, Cambridge University Press 2020.
- Build It, Break It, Fix It: Contesting Secure Development. In ACM Transactions on Privacy and Security (TOPS) 2020.
- Quantifying vulnerability of secret generation using hyper-distributions. In POST 2017.
- Build It, Break It, Fix It: Contesting Secure Development. In CCS 2016.
- Towards the Quantification of Strategy Leakage. In FCS 2016.
- Picking vs. Guessing Secrets: A Game-Theoretic Analysis. In CSF 2015.
- Build It Break It: Measuring and Comparing Development Security. In CSET 2015.
- Quantifying Information Flow for Dynamic Secrets. In S&P/Oakland 2014.
- Adversary Gain vs. Defender Loss in Quantified Information Flow. In FCS 2014.
- Dynamic Enforcement of Knowledge-based Security Policies using Probabilistic Abstract Interpretation. In JCS 2013.
- Knowledge Inference for Optimizing Secure Multi-party Computation. In PLAS 2013.
- Knowledge-Oriented Secure Multiparty Computation. In PLAS 2012.
- Dynamic Enforcement of Knowledge-based Security Policies. In CSF 2011.
Static (and Probabilistic) Analysis
- Probabilistic Abstract Interpretation: Sound Inference and Application to Privacy. In Foundations of Probabilistic Programming, Cambridge University Press 2020.
- Evaluating Design Tradeoffs in Numeric Static Analysis for Java. In ESOP 2018.
- Dynamic Enforcement of Knowledge-based Security Policies using Probabilistic Abstract Interpretation. In JCS 2013.
- Knowledge-Oriented Secure Multiparty Computation. In PLAS 2012.
- Dynamic Enforcement of Knowledge-based Security Policies. In CSF 2011.
Research Community Service
Committees/Reviewer
- AAAI Conference on Artificial Intelligence (AAAI) 2021-2023
- ACL Annual Meeting of the Association for Computational Linguistics (ACL) 2021, 2023, 2024
- ACM Computing Surveys (ACM-CS) ~2017, ~2023-2024
- ACM Transactions on Privacy and Security (ACM-TOPS) ~2020
- ACM Workshop on Artificial Intelligence and Security (AISec) 2017
- ACM ASIA Conference on Computer & Communications Security (ASIACCS) 2014
- ACM Conference on Computer & Communications Security (CCS) 2016-2017, 2020, 2024
- IEEE Computer Security Foundations Symposium (CSF) 2013-2018, 2020, 2024,
- Workshop on Data and Algorithmic Transparency (DAT) 2016
- ACL Empirical Methods in Natural Language Processing (EMNLP) 2023
- Entropy (Entropy) ~2019
- Conference on Fairness, Accountability, and Transparency (FAccT) 2018, 2019, 2021
- Workshop on Fairness, Accountability, and Transparency in Machine Learning (FATML) 2016, 2017
- Workshop on Foundations of Computer Security (FCS) 2016
- EATCS International Colloquium on Automata, Languages, and Programming (ICALP) 2016
- International Conference on Learning Representations (ICLR) 2024
- International Journal of Approximate Reasoning (IJA) ~2017
- Computer and System Sciences (JCSS) ~2015
- Logical Methods in Computer Science (LMCS) ~2017
- International Conference for Military Communications (MILCOM) 2016
- Network and Distributed System Security Symposium (NDSS) 2016-2017
- Neural Information Processing Systems (NeurIPS) 2023
- Privacy Enhancing Technologies Symposium (PETS) 2021
- ACM Workshop on Programming Languages and Analysis of Security (PLAS) 2018, 2019 (co-chair)
- ACM Symposium on Principles of Programming Languages (POPL) 2013, 2015
- ETAPS International Conference on Principles of Security and Trust (POST) 2018
- Workshop on Responsible AI (RAI) 2021 (area chair)
- IEEE Symposium on Security and Privacy (S&P/Oakland) 2015-2020
- Theoretical Computer Science (TCS) ~2017
- Workshop on Privacy in Electronic Society (WPES) 2018, 2020
Other service
- National Science Foundation Panel (NSF) ~2018