paper-with-me

홈 › Papers

Embed Me If You Can: A Geometric Perceptron

2020-06-11 · ICCV 2021 10 · Pavlo Melnyk, Michael Felsberg, Mårten Wadenbäck

Solving geometric tasks involving point clouds by using machine learning is a challenging problem. Standard feed-forward neural networks combine linear or, if the bias parameter is included, affine layers and activation functions. Their geometric modeling is limited, which motivated the prior work introducing the multilayer hypersphere perceptron (MLHP). Its constituent part, i.e., the hypersphere neuron, is obtained by applying a conformal embedding of Euclidean space. By virtue of Clifford algebra, it can be implemented as the Cartesian dot product of inputs and weights. If the embedding is applied in a manner consistent with the dimensionality of the input space geometry, the decision surfaces of the model units become combinations of hyperspheres and make the decision-making process geometrically interpretable for humans. Our extension of the MLHP model, the multilayer geometric perceptron (MLGP), and its respective layer units, i.e., geometric neurons, are consistent with the 3D geometry and provide a geometric handle of the learned coefficients. In particular, the geometric neuron activations are isometric in 3D, which is necessary for rotation and translation equivariance. When classifying the 3D Tetris shapes, we quantitatively show that our model requires no activation function in the hidden layers other than the embedding to outperform the vanilla multilayer perceptron. In the presence of noise in the data, our model is also superior to the MLHP.

📄 PDF Abstract BibTeX arXiv:2006.06507

Code (1)

pavlo-melnyk/mlgp-embedme 공식 구현 pytorch

Tasks

3D geometryDecision Making

Similar Papers 제목 키워드 기반

Polyhedrons and Perceptrons Are Functionally Equivalent

2013-11-05 · Daniel Crespin

Mathematical definitions of polyhedrons and perceptron networks are discussed. The formalization of polyhedrons is done in a rather traditional way. For networks, previously proposed systems are developed. Perceptron net…

Learning from Protein Structure with Geometric Vector Perceptrons

2020-09-03 · ICLR 2021 1 · Bowen Jing, Stephan Eismann, Patricia Suriana, Raphael J. L. Townshend 외

Learning on 3D structures of large biomolecules is emerging as a distinct area in machine learning, but there has yet to emerge a unifying network architecture that simultaneously leverages the graph-structured and geome…

Protein DesignRelational Reasoning

CeRF: Convolutional Neural Radiance Fields for New View Synthesis with Derivatives of Ray Modeling

2023-07-14 · Xiaoyan Yang, Dingbo Lu, Yang Li, Chenhui Li 외

In recent years, novel view synthesis has gained popularity in generating high-fidelity images. While demonstrating superior performance in the task of synthesizing novel views, the majority of these methods are still ba…

Neural RenderingNovel View Synthesis

Learning Bilingual Sentence Embeddings via Autoencoding and Computing Similarities with a Multilayer Perceptron

2019-06-05 · WS 2019 8 · Yunsu Kim, Hendrik Rosendahl, Nick Rossenbach, Jan Rosendahl 외

We propose a novel model architecture and training algorithm to learn bilingual sentence embeddings from a combination of parallel and monolingual data. Our method connects autoencoding and neural machine translation to …

Machine TranslationSentenceSentence EmbeddingsTranslation

Euclidean geometry meets graph, a geometric deep learning perspective

2022-01-17 · ICLR Track Blog 2022 5 · Anonymous

Graph neural networks (GNN) have been an active area of machine learning research to tackle various problems in graph data. Sometimes nodes and edges in a graph can have spatial features, such as 3D coordinates of nodes …

Deep LearningGraph Learning