More Like this

1. Learning constraints from demonstrations with grid and parametric representations

   https://doi.org/10.1177/02783649211035177  

   Chou, Glen; Berenson, Dmitry; Ozay, Necmiye (September 2021, The International Journal of Robotics Research)

   We extend the learning from demonstration paradigm by providing a method for learning unknown constraints shared across tasks, using demonstrations of the tasks, their cost functions, and knowledge of the system dynamics and control constraints. Given safe demonstrations, our method uses hit-and-run sampling to obtain lower cost, and thus unsafe, trajectories. Both safe and unsafe trajectories are used to obtain a consistent representation of the unsafe set via solving an integer program. Our method generalizes across system dynamics and learns a guaranteed subset of the constraint. In addition, by leveraging a known parameterization of the constraint, we modify our method to learn parametric constraints in high dimensions. We also provide theoretical analysis on what subset of the constraint and safe set can be learnable from safe demonstrations. We demonstrate our method on linear and nonlinear system dynamics, show that it can be modified to work with suboptimal demonstrations, and that it can also be used to learn constraints in a feature space. 

   more » « less
2. On Approximate Thompson Sampling with Langevin Algorithms

   Mazumdar, Eric; Pacchiano, Aldo; Ma, Yian; Jordan, Michael; Bartlett, Peter (January 2020, Proceedings of the 37th International Conference on Machine Learning)

   Daumé III, Hal; Singh, Aarti (Ed.)

   Thompson sampling for multi-armed bandit problems is known to enjoy favorable performance in both theory and practice. However, its wider deployment is restricted due to a significant computational limitation: the need for samples from posterior distributions at every iteration. In practice, this limitation is alleviated by making use of approximate sampling methods, yet provably incorporating approximate samples into Thompson Sampling algorithms remains an open problem. In this work we address this by proposing two efficient Langevin MCMC algorithms tailored to Thompson sampling. The resulting approximate Thompson Sampling algorithms are efficiently implementable and provably achieve optimal instance-dependent regret for the Multi-Armed Bandit (MAB) problem. To prove these results we derive novel posterior concentration bounds and MCMC convergence rates for log-concave distributions which may be of independent interest. 

   more » « less
3. Inference for Gaussian Processes with Matern Covariogram on Compact Riemannian Manifolds

   Li, D; Tang, W; Banerjee, S (March 2023, Journal of machine learning research)

   Gaussian processes are widely employed as versatile modelling and predictive tools in spa- tial statistics, functional data analysis, computer modelling and diverse applications of machine learning. They have been widely studied over Euclidean spaces, where they are specified using covariance functions or covariograms for modelling complex dependencies. There is a growing literature on Gaussian processes over Riemannian manifolds in order to develop richer and more flexible inferential frameworks for non-Euclidean data. While numerical approximations through graph representations have been well studied for the Mat´ern covariogram and heat kernel, the behaviour of asymptotic inference on the param- eters of the covariogram has received relatively scant attention. We focus on asymptotic behaviour for Gaussian processes constructed over compact Riemannian manifolds. Build- ing upon a recently introduced Mat´ern covariogram on a compact Riemannian manifold, we employ formal notions and conditions for the equivalence of two Mat´ern Gaussian random measures on compact manifolds to derive the parameter that is identifiable, also known as the microergodic parameter, and formally establish the consistency of the maximum like- lihood estimate and the asymptotic optimality of the best linear unbiased predictor. The circle is studied as a specific example of compact Riemannian manifolds with numerical experiments to illustrate and corroborate the theory 

   more » « less
4. On the Variety and Veracity of Cyber Intrusion Alerts Synthesized by Generative Adversarial Networks

   https://doi.org/10.1145/3394503  

   Sweet, Christopher Ryan; Moskal, Stephen; Yang, Shanchieh (April 2020, ACM Transactions on Management Information Systems)

   Many cyber attack actions can be observed but the observables often exhibit intricate feature dependencies, non-homogeneity, and potential for rare yet critical samples. This work tests the ability to model and synthesize cyber intrusion alerts through Generative Adversarial Networks (GANs), which explore the feature space through reconciling between randomly generated samples and the given data that reflects a mixture of diverse attack behaviors. Through a comprehensive analysis using Jensen-Shannon Divergence (JSD), conditional and joint entropy, and mode drops and additions, we show that the Wasserstein-GAN with Gradient Penalty and Mutual Information (WGAN-GPMI) is more effective in learning to generate realistic alerts than models without Mutual Information constraints. The added Mutual Information constraint pushes the model to explore the feature space more thoroughly and increases the generation of low probability yet critical alert features. By mapping alerts to a set of attack stages it is shown that the output of these low probability alerts has a direct contextual meaning for cyber security analysts. Overall, our results show the promising novel use of GANs to learn from limited yet diverse intrusion alerts to generate synthetic ones that emulate critical dependencies, opening the door to data driven network threat models. 

   more » « less
5. How Children Learn From Others: An Analysis of Selective Word Learning

   https://doi.org/10.1111/cdev.13415  

   Sobel, David_M; Finiasz, Zoe (July 2020, Child Development)

   One way children are remarkable learners is that they learn from others. Critically, children are selective when assessing from whom to learn, particularly in the domain of word learning. We conducted an analysis of children’s selective word learning, reviewing 63 papers on 6,525 participants. Children’s ability to engage in selective word learning appeared to be present in the youngest samples surveyed. Their more metacognitive understanding that epistemic competence indicates reliability or that others are good sources of knowledge has more of a developmental trajectory. We also found that various methodological factors used to assess children influence performance. We conclude with a synthesis of theoretical accounts of how children learn from others. 

   more » « less