computer-vision

Decision Making Under Uncertainty

3260 papers • 126 benchmarks • 313 datasets

This task has no description! Would you like to contribute one?

(Image credit: Papersgraph)

Benchmarks

These leaderboards are used to track progress in decision-making-under-uncertainty

Trend

Dataset

Best Model

Actions

No benchmarks available.

Libraries

i

Use these libraries to find decision-making-under-uncertainty models and implementations

grockious/lcrl

3 papers 50

Datasets

xView3-SAR

Risk-Aware Planning Dataset

Subtasks

Uncertainty Visualization

Most implemented papers

Supplementary Material to: Bayesian Optimization of Risk Measures

Sait Cakmak, Raul Astudillo, P. Frazier, Enlu Zhou•Thu Dec 31 2020

Experiments comparing algorithm run times, and results comparing algorithm run times, are published.

61

Content

0

Paper Graph

Neur2SP: Neural Two-Stage Stochastic Programming

Elias Boutros Khalil, Justin Dumouchelle, R. Patel, Merve Bodur•Thu May 19 2022

Neur2SP is a new method that approximates the expected value function via a neural network to obtain a surrogate model that can be solved more efficiently than the traditional extensive formulation approach, and can be implemented using an off-the-shelf MIP solver.

38 0

Paper Graph

Encoding the Latent Posterior of Bayesian Neural Networks for Uncertainty Quantification

G. Franchi, Andrei Bursuc, Séverine Dubuisson, Emanuel Aldea, I. Bloch•Thu Dec 03 2020

This work aims for efficient deep BNNs amenable to complex computer vision architectures, and achieves this by leveraging variational autoencoders (VAEs) to learn the interaction and the latent distribution of the parameters at each network layer.

32 0

Paper Graph

Natural-Parameter Networks: A Class of Probabilistic Neural Networks

Xingjian Shi, Hao Wang, D. Yeung•Tue Nov 01 2016

A class of probabilistic neural networks, dubbed natural-parameter networks (NPN), is proposed as a novel and lightweight Bayesian treatment of NN, which allows the usage of arbitrary exponential-family distributions to model the weights and neurons.

89 0

Paper Graph

Emulation of physical processes with Emukit

Andrei Paleyes, Neil D. Lawrence, Mark Pullin, Maren Mahsereci, Cliff McCollum, Javier I. González•Sun Oct 24 2021

Emukit is a highly adaptable Python toolkit for enriching decision making under uncertainty that allows users to use state of the art methods including Bayesian optimization, multi-fidelity emulation, experimental design, Bayesian quadrature and sensitivity analysis.

90 0

Paper Graph

Calibrating Deep Convolutional Gaussian Processes

Gia-Lac Tran, Edwin V. Bonilla, J. Cunningham, Pietro Michiardi, M. Filippone•Fri May 25 2018

This paper proposes a novel combination of CNNs with Gaussian processes that considerably outperforms previous approaches on this aspect, while achieving state-of-the-art performance on image classification tasks.

44 0

Paper Graph

Factorized Machine Self-Confidence for Decision-Making Agents.

Brett W. Israelsen, N. Ahmed, Eric W. Frew, D. Lawrence, B. Argrow•Sun Oct 14 2018

Using a factorization based framework for self-confidence assessment, one component of self- confidence, called `solver-quality', is discussed in the context of Markov decision processes for autonomous systems, and a method for assessing solver quality is derived, drawing inspiration from empirical hardness models.

1 0

Paper Graph

A Probabilistic Model of the Bitcoin Blockchain

L. Wynter, Marc Jourdan, Sebastien Blandin, P. Deshpande•Tue Nov 06 2018

This work introduces a complete probabilistic model of the Bitcoin Blockchain, setting the basis for follow-up AI applications on Bitcoin transactions and extends the model to include hidden entity attributes such as the functional category of the associated logical agent.

20 0

Paper Graph

Probabilistic Logic Programming with Beta-Distributed Random Variables

M. Sensoy, Lance M. Kaplan, F. Cerutti, Angelika Kimmig•Wed Sep 19 2018

This work enables aProbLog—a probabilistic logical programming approach—to reason in presence of uncertain probabilities represented as Beta-distributed random variables, which achieves the same performance of state-of-the-art algorithms for highly specified and engineered domains, while simultaneously maintaining the flexibility offered by a ProbLog in handling complex relational domains.

12 0

Paper Graph

LCRL: Certified Policy Synthesis via Logically-Constrained Reinforcement Learning

Mohammadhosein Hasanbeig, D. Kroening, A. Abate•Tue Sep 20 2022

Owing to the LDBA-guided exploration and LCRL model-free architecture, robust performance is observed, which also scales well when compared to standard RL approaches (whenever applicable to LTL specifications).

17 0

Paper Graph

Adding a benchmark result helps the community track progress.