Search for: All records

T3-CIDERS is a train-the-trainer program to increase the adoption of advanced cyberinfrastructure (CI) and data skills into the fabric of research and education in cybersecurity and cyber-related disciplines. T3-CIDERS trains faculty, researchers, and students as “future trainers” (FTs) with hands-on technical and instructional skills to enable more people to effectively leverage CI in cybersecurity. The program includes a series of technical pre-training modules, a weeklong summer institute, ongoing learning engagements conducted over an academic year; it culminates with the FTs conducting locally tailored CI-infused training events at their respective home institutions. Ultimately, T3-CIDERS aims to build a “CI+cybersecurity” community of practice as the cohort continues to practice and teach CI skills in their teaching and research activities. This paper describes the vision and implementation of T3-CIDERS with the first cohort starting in year 2024. Based on the lessons learned through the in-person cohorts, a fully online program will be developed to expand the reach of T3-CIDERS to a broader audience. T3-CIDERS responds to the need to close the CI and data skill gap to meet the increasing challenges in securing the digital world. 

Privacy-preserving Machine Learning as a Service (MLaaS) enables the powerful cloud server to run its well-trained neural model upon the input from resource-limited client, with both of server's model parameters and client's input data protected. While computation efficiency is critical for the practical implementation of privacy-preserving MLaaS and it is inspiring to witness recent advances towards efficiency improvement, there still exists a significant performance gap to real-world applications. In general, state-of-the-art frameworks perform function-wise efficiency optimization based on specific cryptographic primitives. Although it is logical, such independent optimization for each function makes noticeable amount of expensive operations unremovable and misses the opportunity to further accelerate the performance by jointly considering privacy-preserving computation among adjacent functions. As such, we propose COIN: Conjunctive Optimization with Interleaved Nexus, which remodels mainstream computation for each function to conjunctive counterpart for composite function, with a series of united optimization strategies. Specifically, COIN jointly computes a pair of consecutive nonlinear-linear functions in the neural model by reconstructing the intermediates throughout the whole procedure, which not only eliminates the most expensive crypto operations without invoking extra encryption enabler, but also makes the online crypto complexity independent of filter size. Experimentally, COIN demonstrates 11.2x to 29.6x speedup over various function dimensions from modern networks, and 6.4x to 12x speedup on the total computation time when applied in networks with model input from small-scale CIFAR10 to large-scale ImageNet. 

This paper proposes SEER, a novel backdoor detection algorithm for vision-language models, addressing the gap in the literature on multi-modal backdoor detection. While backdoor detection in single-modal models has been well studied, the investigation of such defenses in multi-modal models remains limited. Existing backdoor defense mechanisms cannot be directly applied to multi-modal settings due to their increased complexity and search space explosion. In this paper, we propose to detect backdoors in vision-language models by jointly searching image triggers and malicious target texts in feature space shared by vision and language modalities. Our extensive experiments demonstrate that SEER can achieve over 92% detection rate on backdoor detection in vision-language models in various settings without accessing training data or knowledge of downstream tasks.