no code implementations • 22 May 2024 • Ling Han, Hao Huang, Dustin Scheinost, Mary-Anne Hartley, María Rodríguez Martínez
Effective adaptation to distribution shifts in training data is pivotal for sustaining robustness in neural networks, especially when removing specific biases or outdated information, a process known as machine unlearning.
no code implementations • 5 Jan 2023 • Matthew Rosenblatt, Javid Dadashkarimi, Dustin Scheinost
In the spirit of developing attacks to better understand trustworthiness, we developed two techniques to drastically enhance prediction performance of classifiers with minimal changes to features: 1) general enhancement of prediction performance, and 2) enhancement of a particular method over another.
no code implementations • 2 Jul 2021 • Javid Dadashkarimi, Amin Karbasi, Dustin Scheinost
Being able to map connectomes and derived results between different atlases without additional pre-processing is a crucial step in improving interpretation and generalization between studies that use different atlases.