Discovering physical concepts with neural networks
Explore this paper's citation graph
Summary
This work models a neural network architecture after the human physical reasoning process, which has similarities to representation learning, and applies this method to toy examples to show that the network finds the physically relevant parameters, exploits conservation laws to make predictions, and can help to gain conceptual insights.
- Type
- article
- Published
- 2018-07-27
- Cited by
- 449
- References
- 126
- Access
- Open access
- OpenAlex
- https://openalex.org/W2884775584
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:51865583
Keywords
Artificial neural network, Computer science, Statistical physics, Artificial intelligence, Physics
References
- Equations of Motion from a Data Series
- Introduction to Smooth Manifolds
- Auto-Encoding Variational Bayes
- Distilling Free-Form Natural Laws from Experimental Data
- Approximation capabilities of multilayer feedforward networks
- Genetic programming as a means for programming computers by natural selection
- Simulation as an engine of physical scene understanding
- Projective simulation for artificial intelligence
- Comment on the article "Distilling free-form natural laws from experimental data"
- Deep learning and the information bottleneck principle
- Extracting dynamical equations from experimental data is NP hard.
- Breakdown of Predictability in Gravitational Collapse
- Approximation by superpositions of a sigmoidal function
- Learning a Manifold as an Atlas
- Multilayer feedforward networks are universal approximators
- Automated refinement and inference of analytical models for metabolic networks
- Representation Learning: A Review and New Perspectives
- Long-Term Occupancy Analysis Using Graph-Based Optimisation in Thermal Imagery
- Fast and Accurate Deep Network Learning by Exponential Linear Units (ELUs)
- Galileo: Perceiving Physical Object Properties by Integrating a Physics Engine with Deep Learning
Cited by
- Detecting quantum speedup by quantum walk with convolutional neural networks
- Unveiling phase transitions with machine learning
- Relevant feature extraction for statistical inference
- Neural network learns physical rules for copolymer translocation through amphiphilic barriers
- Learning relevant features for statistical inference
- Adding Intuitive Physics to Neural-Symbolic Capsules Using Interaction Networks
- Explainable Machine Learning for Scientific Insights and Discoveries
- SELFIES: a robust representation of semantically constrained graphs with an example application in chemistry
- Hamiltonian Neural Networks
- Event generation and statistical sampling for physics with deep generative models and a density information buffer
- Learning Programmatically Structured Representations with Perceptor Gradients
- Self-supervised learning with physics-aware neural networks – I. Galaxy model fitting
- Extracting Interpretable Physical Parameters from Spatiotemporal Systems using Unsupervised Learning
- A neural network oracle for quantum nonlocality problems in networks
- Artificial Intelligence Implementations on the Blockchain. Use Cases and Future Applications
- Dimensionality Reduction Applied to Time Response of Linear Systems Using Autoencoders
- Predicting Rare Events in Multiscale Dynamical Systems using Machine Learning
- Emergent Schrödinger equation in an introspective machine learning architecture.
- Vulnerability prediction capability: A comparison between vulnerability discovery models and neural network models
- Artificial intelligence, ideas by statistical mechanics, and affective modulation of information processing
Related papers
- ИСПОЛЬЗОВAНИЕ ПОТЕНЦИAЛA СОЦИAЛЬНЫХ ПAРТНЕРОВ В ПОДГОТОВКЕ БУДУЩИХ ПЕДAГОГОВ
- DETERMINING QUALITY REQUIREMENTS AT THE UNIVERSITIES TO IMPROVE THE QUALITY OF EDUCATION
- Using DataGrid Control to Realize DataBase of Querying in VB6.0
- ESKVS: efficient and secure approach for keyframes-based video summarization framework
- Study and Two Types of Typical Usage of DataGrid Web Server Control
- STKVS: secure technique for keyframes-based video summarization model
- PACWON: A parallelizing compiler for workstations on a network
- SLA Aware Optimized Task Scheduling Model for Faster Execution of Workloads Among Federated Clouds