A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
Filters
Geometric Methods for Sampling, Optimisation, Inference and Adaptive Agents
[article]
2022
arXiv
pre-print
Preserved Fulltext
Other Versions
In this chapter, we identify fundamental geometric structures that underlie the problems of sampling, optimisation, inference and adaptive decision-making. ...
Throughout, we emphasise the rich connections between these fields; e.g., inference draws on sampling and optimisation, and adaptive decision-making assesses decisions by inferring their counterfactual ...
Adaptive agents through active inference The previous sections have established some of the mathematical fundaments of optimisation, sampling and inference. ...
arXiv:2203.10592v3
fatcat:4armb4jklnh4dk4gzk2zqg6tye
Localising In Complex Scenes Using Balanced Adversarial Adaptation
[article]
2020
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
We evaluate our method by adapting representations optimised for indoor Habitat simulated environments (Matterport3D and Replica) to a real-world indoor environment (Active Vision Dataset), showing that ...
Our method exploits the shared geometric similarities between simulation and real-world environments whilst maintaining invariance towards visual discrepancies. ...
Acknowledgments The authors would like to thank the anonymous reviewers for their useful and constructive comments. ...
arXiv:2011.04122v1
fatcat:y4cj2ksfqzguzlrperryxcrxdy
Multi-view pose estimation with mixtures of parts and adaptive viewpoint selection
2018
IET Computer Vision
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2019; you can also visit the original URL.
The file type is application/pdf.
We propose a new method for human pose estimation which leverages information from multiple views to impose a strong prior on articulated pose. ...
Experiments on the HumanEva and UMPM datasets show that the proposed method significantly decreases the estimation error compared to single-view results. ...
We propose an adaptive viewpoint selection mechanism and introduce a binary indicator vector (over parts) that switches on and off geometric and appearance constraints for each part during inference. ...
doi:10.1049/iet-cvi.2017.0146
fatcat:64ahzwzejjdlriqse74k564xua
PRISM: Probabilistic Real-Time Inference in Spatial World Models
[article]
2022
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
We introduce PRISM, a method for real-time filtering in a probabilistic generative model of agent motion and visual perception. ...
Previous approaches either lack uncertainty estimates for the map and agent state, do not run in real-time, do not have a dense scene representation or do not model agent dynamics. ...
Acknowledgments We thank our reviewers for the thoughtful discussion, it helped us to better position our contribution. ...
arXiv:2212.02988v1
fatcat:dperibakbzgwbnkmrghtxzdwsy
Socially Supervised Representation Learning: the Role of Subjectivity in Learning Efficient Representations
[article]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Note that this fulltext copy is not of the "primary" version of this work. The version it corresponds to is:
2021 pre-print arXiv:2109.09390v2 fatcat:3czgxisy2zdl5pjvpmdck3gqhyOther Versions
Altogether, our results demonstrate how communication from subjective perspectives can lead to the acquisition of more abstract representations in multi-agent systems, opening promising perspectives for ...
future research at the intersection of representation learning and emergent communication. ...
systems to naturally infer positive samples. ...
arXiv:2109.09390v3
fatcat:46j2mvmxivglnh6di3uc5w6ywa
A review of source term estimation methods for atmospheric dispersion events using static or mobile sensors
2017
Information Fusion
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2019; you can also visit the original URL.
The file type is application/pdf.
Improvements to sampling techniques such as adaptive sampling or using prior information to generate better inferences could significantly reduce the number of iterations required. ...
An adaptive iterative multiple model ABC sampler was proposed to increase the acceptance rate of the rejection sampler by adaptively generating the proposal distribution for each sample. ...
doi:10.1016/j.inffus.2016.11.010
fatcat:j6il5ptp45dg7pwe5es3c24bta
Tracking and Planning with Spatial World Models
[article]
2022
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
We introduce a method for real-time navigation and tracking with differentiably rendered world models. ...
agent dynamics. ...
The same navigation tasks are used for each noise level and method. ...
arXiv:2201.10335v1
fatcat:rfty3jxmqzfenntfinsx2pvfo4
Adaptive Forgetting Factor Fictitious Play
[article]
2011
arXiv
pre-print
Preserved Fulltext
The Internet Archive has a preservation copy of this work in our general collections.
The file type is application/pdf.
We compare the results of the proposed algorithm with those of stochastic and geometric fictitious play in a simple strategic form game, a vehicle target assignment game and a disaster management problem ...
It is now well known that decentralised optimisation can be formulated as a potential game, and game-theoretical learning algorithms can be used to find an optimum. ...
Tables 5-7 and 8-10 depict the results for adaptive forgetting factor fictitious play and geometric fictitious play respectively. ...
arXiv:1112.2315v1
fatcat:flvyeacn2jhujcehucn7ocmwee
Recent Patents on Computational Intelligence
2010
Recent Patents on Computer Science
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
The patents are categorised by their domain of application: Artificial Neural Networks, Fuzzy Logic and Evolutionary Algorithms and further classified by the application area. ...
It discusses and summarises some of these latest developments in terms of patents. ...
Each agent comprises a genetic algorithm and uses several neural networks for computation. ...
doi:10.2174/1874479610801020110
fatcat:2timwdwx7neutiavmb2aiu3xxe
Recent Patents on Computational Intelligence
2008
Recent Patents on Computer Science
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
The patents are categorised by their domain of application: Artificial Neural Networks, Fuzzy Logic and Evolutionary Algorithms and further classified by the application area. ...
It discusses and summarises some of these latest developments in terms of patents. ...
Each agent comprises a genetic algorithm and uses several neural networks for computation. ...
doi:10.2174/2213275910801020110
fatcat:beze53g2kvb57h6k3jf6icko4a
On Machine Learning and Structure for Mobile Robots
[article]
2018
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Due to recent advances - compute, data, models - the role of learning in autonomous systems has expanded significantly, rendering new applications possible for the first time. ...
Building on this coarse and broad survey of current research, the final sections aim to provide insights into future potentials and challenges as well as the necessity of structure in current practical ...
With respect to this review, I'm particularly thankful to Alex Bewley and Ankur Handa for thorough reading and feedback on earlier drafts. ...
arXiv:1806.06003v1
fatcat:mkekvhmkibdxhdjii4jzu7nkyu
Trust, But Verify: A Survey of Randomized Smoothing Techniques
[article]
2023
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2024; you can also visit the original URL.
The file type is application/pdf.
Additionally, we discuss the challenges of existing methodologies and offer insightful perspectives on potential solutions. ...
This study reviews the theoretical foundations, empirical effectiveness, and applications of randomized smoothing in verifying machine learning classifiers. ...
Thus, they have tried to adapt and overcome both the curse of dimensionality and high inference costs. ...
arXiv:2312.12608v1
fatcat:odu3nzxd5rgeppd5g2gq44kdzi
Tensor product algorithms for inference of contact network from epidemiological data
[article]
2024
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2024; you can also visit the original URL.
The file type is application/pdf.
This enables fast and accurate computation of small probabilities and likelihoods. Numerical simulations demonstrate efficient black-box Bayesian inference of the network. ...
We replace the stochastic simulation with solving the chemical master equation for the probabilities of all network states. ...
This confirms that G ⋆ is at least a local optimum for log L(G), and therefore can be inferred by Bayesian optimisation, assuming the optimisation algorithm manages to converge to it. ...
arXiv:2401.15031v1
fatcat:pczeext2zfey5nhhpkq3efrp6y
Dynamic Opponent Modelling in Fictitious Play
2010
Computer journal
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2010; you can also visit the original URL.
The file type is application/pdf.
We then compared the results of the proposed algorithm with those of stochastic and geometric fictitious play in three different strategic form games: a potential game and two climbing hill games, one ...
We also tested our algorithm in two different distributed optimisation scenarios, a vehicle target assignment game and a disaster management problem. ...
These methods have small communication cost since the agents should exchange once the information which is necessary as input for the pre-planning algorithm. ...
doi:10.1093/comjnl/bxq006
fatcat:2buxmeej6fg2pgm3jiv6n3q5qy
Real-time Mapping of Physical Scene Properties with an Autonomous Robot Experimenter
[article]
2022
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
Neural fields can be trained from scratch to represent the shape and appearance of 3D scenes efficiently. ...
this work, we show that a robot can densely annotate a scene with arbitrary discrete or continuous physical properties via its own fully-autonomous experimental interactions, as it simultaneously scans and ...
We would like to thank Charles Collis at Dyson Technology Ltd. for his support and in enabling the secondment of Iain Haughton from Dyson Technology Ltd. to the Dyson Robotics Lab, Imperial College. ...
arXiv:2210.17325v1
fatcat:jcpshe7tvvg3llyyuup46v3k7e
« Previous
Showing results 1 — 15 out of 4,181 results