About

Chris Clifton is a Professor of Statistics at Purdue University. His research interests include Data Mining, Data Science, Machine Learning, and Statistical Disclosure Limitation. He holds a B.S. in Computer Science and Engineering from the Massachusetts Institute of Technology, earned in 1986, and an M.S. in Electrical Engineering and Computer Science from MIT, also in 1986. He completed his Ph.D. in Computer Science at Princeton University in 1991. Clifton has been recognized with numerous awards, including being named an IEEE Fellow in 2020 and an ACM Distinguished Member in 2017. His contributions to the field have been acknowledged through awards such as the IEEE International Conference on Data Mining Outstanding Service Award in 2011, and the College of Science Graduate Student Mentoring Award in 2009. He is actively involved in research and service within the Department of Statistics at Purdue University.

Research topics

• Artificial Intelligence
• Computer Science
• Machine Learning
• Data Mining
• Statistics
• Algorithm
• Engineering
• Mathematics

Selected publications

• Real-Time Programmable Nonlinear Wavefront Shaping with Si Metasurface Driven by Genetic Algorithm

  Engineering · 2025-05-02 · 1 citations

  articleOpen access

  Nonlinear wavefront shaping is crucial for advancing optical technologies, enabling applications in optical computation, information processing, and imaging. However, a significant challenge is that once a metasurface is fabricated, the nonlinear wavefront it generates is fixed, offering little flexibility. This limitation often necessitates the fabrication of different metasurfaces for different wavefronts, which is both time-consuming and inefficient. To address this, we combine evolutionary algorithms with spatial light modulators (SLMs) to dynamically control wavefronts using a single metasurface, reducing the need for multiple fabrications and enabling the generation of arbitrary nonlinear wavefront patterns without requiring complicated optical alignment. We demonstrate this approach by introducing a genetic algorithm (GA) to manipulate visible wavefronts converted from near-infrared light via third-harmonic generation (THG) in a silicon metasurface. The Si metasurface supports multipolar Mie resonances that strongly enhance light-matter interactions, thereby significantly boosting THG emission at resonant positions. Additionally, the cubic relationship between THG emission and the infrared input reduces noise in the diffractive patterns produced by the SLM. This allows for precise experimental engineering of the nonlinear emission patterns with fewer alignment constraints. Our approach paves the way for self-optimized nonlinear wavefront shaping, advancing optical computation and information processing techniques.

  PublisherDOI

• Real-Time Access Control for Background and Co-Occurrence Image Privacy Protection

  IEEE Transactions on Emerging Topics in Computing · 2025-05-28

  articleSenior author

  In today's digital age, the proliferation of social networks and advanced camera technology has led to countless images being shared on online social platforms daily, potentially resulting in significant breaches of personal privacy. In recent years, many methods have been proposed to protect image privacy, allowing users to be notified of potential privacy leaks before publishing their photos. However, most existing research primarily addresses the privacy protection of image owners or co-owners, while neglecting the privacy of people who appear in the background of others' images or who are co-occurring with others in the same image. In this paper, we propose a system capable of conducting real-time access control for protecting privacy of every individual appearing in a photo, as well as the privacy of people who co-occur in the same image. Specifically, we first detect all the faces in the image, then use a facial recognition algorithm to identify the corresponding users' privacy policies, and finally determine whether the image violates any user's privacy policy. In order to provide real-time access control, we have designed a facial attribute index tree to speed up the process of user identification. The experimental results show that compared with the method without our proposed index tree, our approach improves the time efficiency by almost two orders of magnitude while maintaining the accuracy of more than 97%.

  PublisherDOI

• Welcome to IEEE Transactions on Privacy!

  IEEE Transactions on Privacy · 2024-01-01

  articleOpen access1st authorCorresponding
  PublisherDOI

• On Improving Fairness of AI Models with Synthetic Minority Oversampling Techniques

  Society for Industrial and Applied Mathematics eBooks · 2023 · 8 citations

  Senior authorCorresponding
  • Computer Science
  • Machine Learning
  • Artificial Intelligence

  Biased AI models result in unfair decisions. In response, a number of algorithmic solutions have been engineered to mitigate bias, among which the Synthetic Minority Oversampling Technique (SMOTE) has been studied, to an extent. Although the SMOTE technique and its variants have great potentials to help improve fairness, there is little theoretical justification for its success. In addition, formal error and fairness bounds are not clearly given. This paper attempts to address both issues. We prove and demonstrate that synthetic data generated by oversampling underrepresented groups can mitigate algorithmic bias in AI models, while keeping the predictive errors bounded. We further compare this technique to the existing state-of-the-art fair AI techniques on five datasets using a variety of fairness metrics. We show that this approach can effectively improve fairness even when there is a significant amount of label and selection bias, regardless of the baseline AI algorithm.

  PublisherDOI

• LuMaMi28: Real-Time Millimeter-Wave Multi-User MIMO Systems With Antenna Selection

  IEEE Transactions on Wireless Communications · 2023-03-28 · 5 citations

  article

  This paper presents LuMaMi28, a real-time 28 GHz multi-user (MU) multiple-input multiple-output (MIMO) testbed. In this testbed, the base station has 16 transceiver chains with a fully-digital beamforming architecture (with different pre-coding algorithms) and simultaneously supports multiple user equipments (UEs) with spatial multiplexing. The UEs are equipped with a beam-switchable antenna array for real-time antenna selection where the one with the highest channel magnitude, out of four pre-defined beams, is selected. For the beam-switchable antenna array, we consider two kinds of UE antennas, with different beam-width and different peak-gain. Based on this testbed, we provide measurement results for millimeter-wave (mmWave) MU-MIMO performance in different real-life scenarios with static and mobile UEs. We explore the potential benefit of the mmWave MU-MIMO systems with antenna selection based on measured channel data, and discuss the performance results through real-time measurements.

  PublisherDOI

• Differentially Private <i>k</i> -Nearest Neighbor Missing Data Imputation

  ACM Transactions on Privacy and Security · 2022-03-29 · 8 citations

  article1st authorCorresponding

  Using techniques employing smooth sensitivity , we develop a method for \( k \) -nearest neighbor missing data imputation with differential privacy. This requires bounding the number of data incomplete tuples that can have their data complete “donor” changed by making a single addition or deletion to the dataset. The multiplicity of a single individual’s impact on an imputed dataset necessarily means our mechanisms require the addition of more noise than mechanisms that ignore missing data, but we show empirically that this is significantly outweighed by the bias reduction from imputing missing data.

  PublisherDOI

• A Roadmap for Greater Public Use of Privacy-Sensitive Government Data: Workshop Report

  arXiv (Cornell University) · 2022-06-17 · 1 citations

  preprintOpen access1st authorCorresponding

  Government agencies collect and manage a wide range of ever-growing datasets. While such data has the potential to support research and evidence-based policy making, there are concerns that the dissemination of such data could infringe upon the privacy of the individuals (or organizations) from whom such data was collected. To appraise the current state of data sharing, as well as learn about opportunities for stimulating such sharing at a faster pace, a virtual workshop was held on May 21st and 26th, 2021, sponsored by the National Science Foundation (NSF) and National Institute of Standards and Technologies (NIST), and the White House Office of Science and Technology Policy (OSTP), where a multinational collection of researchers and practitioners were brought together to discuss their experiences and learn about recently developed technologies for managing privacy while sharing data. The workshop specifically focused on challenges and successes in government data sharing at various levels. The first day focused on successful examples of new technology applied to sharing of public data, including formal privacy techniques, synthetic data, and cryptographic approaches. Day two emphasized brainstorming sessions on some of the challenges and directions to address them.

  PublisherOA PDFDOI

• Unfair AI: It Isn’t Just Biased Data

  2022 IEEE International Conference on Data Mining (ICDM) · 2022-11-01 · 5 citations

  article

  Conventional wisdom holds that discrimination in machine learning is a result of historical discrimination: biased training data leads to biased models. We show that the reality is more nuanced; machine learning can be expected to induce types of bias not found in the training data. In particular, if different groups have different optimal models, and the optimal model for one group has higher accuracy, the optimal accuracy joint model will induce disparate impact even when the training data does not display disparate impact. We argue that due to systemic bias, this is a likely situation, and simply ensuring training data appears unbiased is insufficient to ensure fair machine learning.

  PublisherDOI

• LuMaMi28: Real-Time Millimeter-Wave Massive MIMO Systems with Antenna Selection

  arXiv (Cornell University) · 2021-09-07 · 1 citations

  preprintOpen access

  This paper presents LuMaMi28, a real-time 28 GHz massive multiple-input multiple-output (MIMO) testbed. In this testbed, the base station has 16 transceiver chains with a fully-digital beamforming architecture (with different pre-coding algorithms) and simultaneously supports multiple user equipments (UEs) with spatial multiplexing. The UEs are equipped with a beam-switchable antenna array for real-time antenna selection where the one with the highest channel magnitude, out of four pre-defined beams, is selected. For the beam-switchable antenna array, we consider two kinds of UE antennas, with different beam-width and different peak-gain. Based on this testbed, we provide measurement results for millimeter-wave (mmWave) massive MIMO performance in different real-life scenarios with static and mobile UEs. We explore the potential benefit of the mmWave massive MIMO systems with antenna selection based on measured channel data, and discuss the performance results through real-time measurements.

  PublisherOA PDFDOI

• Differentially Private Naïve Bayes Classifier Using Smooth Sensitivity

  Proceedings on Privacy Enhancing Technologies · 2021-07-23 · 1 citations

  preprintOpen accessSenior author

  Abstract There is increasing awareness of the need to protect individual privacy in the training data used to develop machine learning models. Differential Privacy is a strong concept of protecting individuals. Naïve Bayes is a popular machine learning algorithm, used as a baseline for many tasks. In this work, we have provided a differentially private Naïve Bayes classifier that adds noise proportional to the smooth sensitivity of its parameters. We compare our results to Vaidya, Shafiq, Basu, and Hong [1] which scales noise to the global sensitivity of the parameters. Our experimental results on real-world datasets show that smooth sensitivity significantly improves accuracy while still guaranteeing ɛ -differential privacy.

  PublisherOA PDFDOI

Recent grants

• ITR - (ASE+NHS) - (dmc+int): Privacy-Preserving Data Integration and Sharing

  NSF · $1.0M · 2004–2008

• Shared Software Database

  NSF · $90k · 1992–1995

Frequent coauthors

• Murat Kantarcıoğlu

  The University of Texas at Dallas

  24 shared

• Jaideep Vaidya

  24 shared

• Wen‐Syan Li

  Seoul National University

  14 shared

• Koray Mancuhan

  10 shared

• Mehmet Ercan Nergiz

  Teknoloji Arastirma ve Gelistirme Endustriyel Urunler Bilisim Teknolojileri San Tic

  9 shared

• Wei Jiang

  Soochow University

  8 shared

• Bhavani Thuraisingham

  8 shared

• M. Murugesan

  Sri Sivasubramaniya Nadar College of Engineering

  7 shared

Education

• B.S., Computer Science and Engineering

  Massachusetts Institute of Technology

  1986

• M.S., Electrical Engineering and Computer Science

  Massachusetts Institute of Technology

  1986

• M.A., Computer Science

  Princeton University

  1988

• Ph.D., Computer Science

  Princeton University

  1991

Awards & honors

• IEEE Fellow (2020)
• ACM Distinguished Member (2017)
• College of Science Team Award (2011)
• IEEE International Conference on Data Mining Outstanding Ser…
• Teaching for Tomorrow Award (2011)