Filters








299,675 Hits in 4.4 sec

Learning to Linearize Under Uncertainty [article]

Ross Goroshin, Michael Mathieu, Yann LeCun
2015 arXiv   pre-print
Training deep feature hierarchies to solve supervised learning tasks has achieved state of the art performance on many problems in computer vision.  ...  In this work we suggest a new architecture and loss for training deep feature hierarchies that linearize the transformations observed in unlabeled natural video sequences.  ...  Subsection 3.2 introduces a latent variable formulation as a means of learning to linearize under uncertainty.  ... 
arXiv:1506.03011v2 fatcat:uxpohexffzdxpd6ryijre2oviq

From Predictions to Decisions: Using Lookahead Regularization [article]

Nir Rosenfeld, Sophie Hilgard, Sai Srivatsa Ravindranath, David C. Parkes
2020 arXiv   pre-print
The standard approach to learning is agnostic to induced user actions and provides no guarantees as to the effect of actions.  ...  This regularization carefully tailors the uncertainty estimates governing confidence in this improvement to the distribution of model-induced actions.  ...  In particular, we are interested in learning uncertainty intervals that provide good coverage. There are many approaches to learning under covariate shift.  ... 
arXiv:2006.11638v2 fatcat:gcnmhr5r4nf63pgjzu2fdg4o3u

Reasoning under Uncertainty with Log-Linear Description Logics

Mathias Niepert
2011 International Semantic Web Conference  
We demonstrate the ways in which log-linear description logics answer to these requirements.  ...  The position paper provides a brief summary of log-linear description logics and their applications.  ...  It is therefore hard to imagine how the Semantic Web could succeed without the ability to represent and reason under uncertainty.  ... 
dblp:conf/semweb/Niepert11 fatcat:vytj4gawxfhpxpcbwo2b323iv4

Managing disinflation under uncertainty

M.F. Tesfaselassie, E. Schaling
2010 Journal of Economic Dynamics and Control  
We derive the optimal policy under learning (DOP) and compare it two limiting cases---certainty equivalence policy (CEP) and cautionary policy (CP).  ...  We derive the optimal policy under learning (DOP) and compare it two limiting cases-certainty equivalence policy (CEP) and cautionary policy (CP).  ...  The certainty equivalence policy and the cautionary policy ignore the non-linear updating equations, and so policy is conducted under passive learning and the policy rules are linear in the state variable  ... 
doi:10.1016/j.jedc.2010.07.005 fatcat:z7jnghfeknc7najsf6wygf3syy

Robust PD-Type Iterative Learning Control of Discrete Linear Repetitive Processes in the Finite Frequency Domain

Lei Wang, Mu Li, and Huizhong Yang
2020 Mathematics  
This paper studies a robust iterative learning control design for discrete linear repetitive processes in the finite frequency domain.  ...  The robust control problem with norm-bounded uncertainty and convex polyhedral uncertainty are also considered in this paper.  ...  In order to solve the actuator failure problem of discrete linear repetitive processes with convex polyhedral uncertainties, a robust fault-tolerant iterative learning controller is designed by using the  ... 
doi:10.3390/math8061004 fatcat:y3xaibsphrad3ho4aepw4k3rcu

Reinforcement Learning for Safety-Critical Control under Model Uncertainty, using Control Lyapunov Functions and Control Barrier Functions [article]

Jason Choi, Fernando Castañeda, Claire J. Tomlin, Koushil Sreenath
2020 arXiv   pre-print
Specifically, we propose a novel reinforcement learning framework which learns the model uncertainty present in the CBF and CLF constraints, as well as other control-affine dynamic constraints in the quadratic  ...  model uncertainty.  ...  [22] proposes an RLbased method to learn the model uncertainty compensation for input-output linearization control. In Castañeda et al.  ... 
arXiv:2004.07584v2 fatcat:xhzqmq4xybahxjz4342pg5ibsq

Extremum seeking-based iterative learning linear MPC

Mouhacine Benosman, Stefano Di Cairano, Avishai Weiss
2014 2014 IEEE Conference on Control Applications (CCA)  
We assume linear models with parametric uncertainties, and propose an iterative multi-variable extremum seeking (MES)-based learning MPC algorithm to learn on-line the uncertain parameters and update the  ...  In this work we study the problem of adaptive MPC for linear time-invariant uncertain models.  ...  Control objective We want to design an adaptive controller that solves regulation and tracking problems for linear time-invariant systems with structural model uncertainties under state, input, and output  ... 
doi:10.1109/cca.2014.6981582 dblp:conf/IEEEcca/BenosmanCW14 fatcat:tystq22x75hktmzr5orlm7d7je

Robust iterative learning control with current feedback for uncertain linear systems

Tae-Yong Doh
1999 International Journal of Systems Science  
linear fractional transformations (L FTs).  ...  In this method, a feedback controller and learning controllers can be designed at one time and a weighting function is introduced to increase the learning performance.  ...  De Roover (1996) synthesized an iterative learning controller based on control under unstructured uncertainty.  ... 
doi:10.1080/002077299292650 fatcat:5bhv6li3ujf2xdbiouw6dcilca

Baseline Methods for Active Learning

Gavin C. Cawley
2011 Journal of machine learning research  
In this paper, we describe some simple pool-based active learning strategies, based on optimally regularised linear [kernel] ridge regression, providing a set of baseline submissions for the Active Learning  ...  This motivates the development of active learning methods, that seek to direct the collection of labelled examples such that the greatest performance gains can be achieved using the smallest quantity of  ...  Acknowledgments I would like to thank the anonymous reviewers for their helpful and constructive comments and the co-organizers of the challenge for their efforts in staging a very interesting and (for  ... 
dblp:journals/jmlr/Cawley11 fatcat:4srykoko4jcjtlndaapdonxguq

A framework for benchmarking uncertainty in deep regression [article]

Franko Schmähling, Jörg Martin, Clemens Elster
2021 arXiv   pre-print
We illustrate the proposed framework by applying it to current approaches for uncertainty quantification in deep regression.  ...  The framework is based on regression problems where the regression function is a linear combination of nonlinear functions.  ...  In the out-of-distribution range the uncertainty grows for the deep learning methods, but is often to small to cover the ground truth.  ... 
arXiv:2109.09048v1 fatcat:i3s7f4dqwzgbffdt7uolihuvme

Uncertainty Quantification in Neural Differential Equations [article]

Olga Graf, Pablo Flores, Pavlos Protopapas, Karim Pichara
2021 arXiv   pre-print
Uncertainty quantification (UQ) helps to make trustworthy predictions based on collected observations and uncertain domain knowledge.  ...  Among applications that can benefit from effective handling of uncertainty are the deep learning based differential equation (DE) solvers.  ...  ., calibrating ε with residuals at each optimization step) will help the deep learning based DE solutions to outperform classical ones and lead to their increased presence in applications.  ... 
arXiv:2111.04207v1 fatcat:iscpc637lfahlnbgvhi5akjcfu

Safe Learning in Robotics: From Learning-Based Control to Safe Reinforcement Learning [article]

Lukas Brunke, Melissa Greeff, Adam W. Hall, Zhaocong Yuan, Siqi Zhou, Jacopo Panerati, Angela P. Schoellig
2021 arXiv   pre-print
This article provides a concise but holistic review of the recent advances made in using machine learning to achieve safe decision making under uncertainties, with a focus on unifying the language and  ...  As data- and learning-based robot control methods continue to gain traction, researchers must understand when and how to best leverage them in real-world scenarios where safety is imperative, such as when  ...  ACKNOWLEDGMENTS The authors would like to acknowledge the early contributions to this work by Karime Pereida and Sepehr Samavi, the invaluable suggestions and feedback by Hallie Siegel, as well as the  ... 
arXiv:2108.06266v2 fatcat:gbbe3qyatfgelgzhqzglecr5qm

MONETARY POLICY RULES UNDER UNCERTAINTY: EMPIRICAL EVIDENCE, ADAPTIVE LEARNING, AND ROBUST CONTROL

WENLANG ZHANG, WILLI SEMMLER
2005 Macroeconomic Dynamics  
Facing such uncertainties, a central bank may resort to different strategies, it can either reduce uncertainty by learning or just choose a policy rule robust to uncertainty.  ...  This paper explores monetary policy rules under model and shock uncertainties.  ...  We consider both linear and nonlinear Phillips curves. In Section 3 we explore monetary policy rules under model uncertainty with adaptive learning.  ... 
doi:10.1017/s1365100505040332 fatcat:dtn5czy4y5exlfxpvfuqknd4c4

Non-Linear Spectral Dimensionality Reduction Under Uncertainty [article]

Firas Laakom, Jenni Raitoharju, Nikolaos Passalis, Alexandros Iosifidis, Moncef Gabbouj
2022 arXiv   pre-print
In this paper, we consider the problem of non-linear dimensionality reduction under uncertainty, both from a theoretical and algorithmic perspectives.  ...  Since real-world data usually contain measurements with uncertainties and artifacts, the input space in the proposed framework consists of probability distributions to model the uncertainties associated  ...  GEU is able to learn only linear projections and, thus, it fails if non-linearity is required as in the case of MFA-GE.  ... 
arXiv:2202.04678v1 fatcat:cpbdiohmbvhelp7efmq5xakam4

Theoretical framework for managing the front end of innovation under uncertainty

Richard Sperry, Antonie Jetter
2009 PICMET '09 - 2009 Portland International Conference on Management of Engineering & Technology  
Instead, projects with different market and technical uncertainties should be managed with one of five different processes (linear, recursive, evolving, selectionism, trial-and-error).  ...  FRAMEWORK FOR MANAGING THE FRONT END UNDER UNCERTAINTY TABLE 6 : 6 FUZZY FRONT END PROCESSES BY UNCERTAINTIES Suitable Innovation Project Based on Performance Uncertainties FFE Processes  ...  Research has characterized the linear process to be more suitable for incremental improvement [9, 18, 20, 33] with medium-low to low levels of uncertainty. IV.  ... 
doi:10.1109/picmet.2009.5261940 fatcat:dsdxhj6b45d4tlizre7efuhc7q
« Previous Showing results 1 — 15 out of 299,675 results