machine-learning

Democratizing Health Research Through Privacy-Protecting Synthetic Data

This project aims to enable significantly broader use of health data by creating differentially private synthetic data sets. This project will also contribute to solutions for the focus on the coronavirus pandemic. It is supported by the UC Davis CeDAR.

Provable Anonymization of Grid Data for Cyberattack Detection

This project aims to develop techniques for enabling data analysis for the purposes of detecting and/or investigating cyberattacks against energy delivery systems while also preserving aspects of key confidentiality elements within the underlying raw data being analyzed. The result will be a complete solution for anonymization of data collected from OT and IT networks pertaining to energy grid cyberattack detection that has been tested for its ability to retain privacy properties and still enable attack detection. It is funded by DOE CESER’s CEDS program and is led by Sean Peisert.

Supervisory Parameter Adjustment for Distribution Energy Storage (SPADES)

This project is developing the methodology and tools allowing Electric Storage Systems (ESS) to automatically reconfigure themselves to counteract cyberattacks, both directly against the ESS control systems and indirectly through the electric grid. It is funded by DOE CESER’s CEDS program and is led by Daniel Arnold.

Privacy-Preserving Data Analysis for Scientific Discovery

This project aims to produce methods, processes, and architectures applicable to a variety of scientific computing domains that enables querying, machine learning, and analysis of data while protecting against releasing sensitive information beyond pre-defined bounds. It is supported by LBNL CSR funds and is led by Sean Peisert.

Cybersecurity via Inverter-Grid Automatic Reconfiguration (CIGAR)

This project is performing R&D to enable distribution grids to adapt to resist a cyber-attack by (1) developing adaptive control algorithms for DER, voltage regulation, and protection systems; (2) analyze new attack scenarios and develop associated defensive strategies. It is funded by DOE CESER’s CEDS program and is co-led by Sean Peisert and Daniel Arnold.

Integrated Multi Scale Machine Learning for the Power Grid

The goal of this project is to create advanced, distributed data analytics capability to provide visibility and controllability to distribution grid operators. It is funded by the DOE Grid Modernization Initiative. The LBNL portion of this effort is led by Sean Peisert.

Detecting Distributed Denial of Service Attacks on Wide-Area Networks

This project develops techniques for detecting DDoS attacks and disambiguating them from large-scale science flows. It is funded by the DOE iJC3 Cyber R&D program and is led by Sean Peisert.

Toward a Hardware/Software Co-Design Framework for Ensuring the Integrity of Exascale Scientific Data

This project takes a broad look at several aspects of security and scientific integrity issues in HPC systems. It is funded by DOE ASCR and is led by Sean Peisert.

Power Grid Threat Detection and Response with Data Analytics

The goal of this project is to develop technologies and methodologies to protect the nation’s power grid from advanced cyber and all-hazard threats. This will be done through the collection of disparate data and the use of advanced analytics to detect threats and response to them. It is funded by DOE OE’s CEDS program via the Grid Modernization Initiative and is co-led by Sean Peisert.

Inferring Computing Activity Using Physical Sensors

This project uses power data to monitor the use of computing systems, including supercomputers and large computing centers. It is led by Sean Peisert.

Secure and Private Acquisition, Storage, and Analysis of Medical Sensor Data

This project is developing a system-based workflow to securely acquire wireless data from mechanical ventilators in critical care environments, and leverage scalable web-based analytic platforms to advance data analytics and visualization of issues surrounding patients with respiratory failure.

NetSage - an open privacy-aware network measurement, analysis, and visualization service

NetSage is a network measurement, analysis and visualization service funded by the National Science Foundation and is designed to address the needs of today’s international networks. This project is co-led by Sean Peisert at LBNL.

Cyber Security of Power Distribution Systems by Detecting Differences Between Real-time Micro-Synchrophasor Measurements and Cyber-Reported SCADA

This project is using micro-PMU measurements and SCADA commands to develop a system to detect cyberattacks against the power distribution grid. It is funded by DOE OE’s CEDS program and is led by Sean Peisert.

The Hive Mind: Applying a Distributed Security Sensor Network to GENI.

This project sought to define and prototype a security layer using a method of intrusion detection based on mobile agents and swarm intelligence. The project was funded by NSF’s CISE Directorate, and was led by Sean Peisert.

A Mathematical and Data-Driven Approach to Intrusion Detection for High-Performance Computing

This project developed mathematical and statistical techniques to analyze the secure access and use of high-performance computer systems. It was funded by DOE ASCR and was originally led by David H. Bailey.