Unconstrained scene generation with locally conditioned radiance fields

T DeVries, MA Bautista, N Srivastava… - Proceedings of the …, 2021 - openaccess.thecvf.com
We tackle the challenge of learning a distribution over complex, realistic, indoor scenes. In
this paper, we introduce Generative Scene Networks (GSN), which learns to decompose …

Sha** belief states with generative environment models for rl

K Gregor, D Jimenez Rezende… - Advances in …, 2019 - proceedings.neurips.cc
When agents interact with a complex environment, they must form and maintain beliefs
about the relevant aspects of that environment. We propose a way to efficiently train …

Graph element networks: adaptive, structured computation and memory

F Alet, AK Jeewajee, MB Villalonga… - International …, 2019 - proceedings.mlr.press
We explore the use of graph neural networks (GNNs) to model spatial processes in which
there is no a priori graphical structure. Similar to finite element analysis, we assign nodes of …

Sequential neural processes

G Singh, J Yoon, Y Son, S Ahn - Advances in Neural …, 2019 - proceedings.neurips.cc
Neural Processes combine the strengths of neural networks and Gaussian processes to
achieve both flexible learning and fast prediction in stochastic processes. However, a large …

On the link between conscious function and general intelligence in humans and machines

A Juliani, K Arulkumaran, S Sasai, R Kanai - ar**_Unsupervised_Map_Estimation_From_Multiple_Point_Clouds_CVPR_2019_paper.pdf" data-clk="hl=bg&sa=T&oi=gga&ct=gga&cd=5&d=9988954344653969501&ei=vDXCZ6e9GZeY6rQPjuS70As" data-clk-atid="Xdixfg3ln4oJ" target="_blank">[PDF] thecvf.com

DeepMap**: Unsupervised map estimation from multiple point clouds

L Ding, C Feng - Proceedings of the IEEE/CVF conference …, 2019 - openaccess.thecvf.com
We propose DeepMap**, a novel registration framework using deep neural networks
(DNNs) as auxiliary functions to align multiple point clouds from scratch to a globally …

Generalisation of structural knowledge in the hippocampal-entorhinal system

J Whittington, T Muller, S Mark… - Advances in neural …, 2018 - proceedings.neurips.cc
A central problem to understanding intelligence is the concept of generalisation. This allows
previously learnt structure to be exploited to solve tasks in novel situations differing in their …

Recurrent attentive neural process for sequential data

S Qin, J Zhu, J Qin, W Wang, D Zhao - arxiv preprint arxiv:1910.09323, 2019 - arxiv.org
Neural processes (NPs) learn stochastic processes and predict the distribution of target
output adaptively conditioned on a context set of observed input-output pairs. Furthermore …

Spatially-Aware Transformer for Embodied Agents

J Cho, J Yoon, S Ahn - arxiv preprint arxiv:2402.15160, 2024 - arxiv.org
Episodic memory plays a crucial role in various cognitive processes, such as the ability to
mentally recall past events. While cognitive science emphasizes the significance of spatial …

Deep latent variable models for sequential data

M Fraccaro - 2018 - orbit.dtu.dk
This thesis was prepared at the Cognitive Systems section of DTU Compute, Department of
Applied Mathematics and Computer Science, Technical University of Denmark. It constitutes …