no code implementations • 26 Oct 2022 • Rouzbeh Behnia, Mohamamdreza Ebrahimi, Jason Pacheco, Balaji Padmanabhan
Differential privacy (DP) provides a rigorous framework that allows adding noise in the process of training or fine-tuning LLMs such that extracting the training data becomes infeasible (i. e., with a cryptographically small success probability).
no code implementations • 27 Jan 2022 • Xiwen Chen, Hao Wang, Abolfazl Razi, Brendan Russo, Jason Pacheco, John Roberts, Jeffrey Wishart, Larry Head, Alonso Granados Baca
To bridge these two perspectives, we define a new set of network-level safety metrics (NSM) to assess the overall safety profile of traffic flow by processing imagery taken by RSU cameras.
no code implementations • NeurIPS 2020 • Sue Zheng, David Hayden, Jason Pacheco, John W. Fisher III
As a result, many algorithms utilize MI bounds as proxies that lack regret-style guarantees.
no code implementations • 20 Nov 2020 • Christopher L. Dean, Stephen J. Lee, Jason Pacheco, John W. Fisher III
We present an approach to data fusion that combines the interpretability of structured probabilistic graphical models with the flexibility of neural networks.
no code implementations • ICML 2018 • Sue Zheng, Jason Pacheco, John Fisher
In many sequential planning applications a natural approach to generating high quality plans is to maximize an information reward such as mutual information (MI).
no code implementations • NeurIPS 2017 • Daniel Milstein, Jason Pacheco, Leigh Hochberg, John D. Simeral, Beata Jarosiewicz, Erik Sudderth
We propose a dynamic Bayesian network that includes the on-screen goal position as part of its latent state, and thus allows the person’s intended angle of movement to be aggregated over a much longer history of neural activity.
no code implementations • NeurIPS 2012 • Jason Pacheco, Erik B. Sudderth
We develop convergent minimization algorithms for Bethe variational approximations which explicitly constrain marginal estimates to families of valid distributions.