Autonomous Learning of Generative Models with Chemical Reaction Network
Ensembles
- URL: http://arxiv.org/abs/2311.00975v2
- Date: Mon, 6 Nov 2023 19:07:59 GMT
- Title: Autonomous Learning of Generative Models with Chemical Reaction Network
Ensembles
- Authors: William Poole, Thomas E. Ouldridge, and Manoj Gopalkrishnan
- Abstract summary: We develop a general architecture whereby a broad class of chemical systems can autonomously learn complex distributions.
Our construction takes the form of a chemical implementation of machine learning's optimization workhorse: gradient descent on the relative entropy cost function.
- Score: 0.0
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: Can a micron sized sack of interacting molecules autonomously learn an
internal model of a complex and fluctuating environment? We draw insights from
control theory, machine learning theory, chemical reaction network theory, and
statistical physics to develop a general architecture whereby a broad class of
chemical systems can autonomously learn complex distributions. Our construction
takes the form of a chemical implementation of machine learning's optimization
workhorse: gradient descent on the relative entropy cost function. We show how
this method can be applied to optimize any detailed balanced chemical reaction
network and that the construction is capable of using hidden units to learn
complex distributions. This result is then recast as a form of integral
feedback control. Finally, due to our use of an explicit physical model of
learning, we are able to derive thermodynamic costs and trade-offs associated
to this process.
Related papers
- Machine learning of hidden variables in multiscale fluid simulation [77.34726150561087]
Solving fluid dynamics equations often requires the use of closure relations that account for missing microphysics.
In our study, a partial differential equation simulator that is end-to-end differentiable is used to train judiciously placed neural networks.
We show that this method enables an equation based approach to reproduce non-linear, large Knudsen number plasma physics.
arXiv Detail & Related papers (2023-06-19T06:02:53Z) - Towards Predicting Equilibrium Distributions for Molecular Systems with
Deep Learning [60.02391969049972]
We introduce a novel deep learning framework, called Distributional Graphormer (DiG), in an attempt to predict the equilibrium distribution of molecular systems.
DiG employs deep neural networks to transform a simple distribution towards the equilibrium distribution, conditioned on a descriptor of a molecular system.
arXiv Detail & Related papers (2023-06-08T17:12:08Z) - Differentiable Programming of Chemical Reaction Networks [63.948465205530916]
Chemical reaction networks are one of the most fundamental computational substrates used by nature.
We study well-mixed single-chamber systems, as well as systems with multiple chambers separated by membranes.
We demonstrate that differentiable optimisation, combined with proper regularisation, can discover non-trivial sparse reaction networks.
arXiv Detail & Related papers (2023-02-06T11:41:14Z) - Implicit Geometry and Interaction Embeddings Improve Few-Shot Molecular
Property Prediction [53.06671763877109]
We develop molecular embeddings that encode complex molecular characteristics to improve the performance of few-shot molecular property prediction.
Our approach leverages large amounts of synthetic data, namely the results of molecular docking calculations.
On multiple molecular property prediction benchmarks, training from the embedding space substantially improves Multi-Task, MAML, and Prototypical Network few-shot learning performance.
arXiv Detail & Related papers (2023-02-04T01:32:40Z) - Port-metriplectic neural networks: thermodynamics-informed machine
learning of complex physical systems [0.09332987715848712]
We develop inductive biases for the machine learning of complex physical systems based on the port-Hamiltonian formalism.
We show that the constructed networks are able to learn the physics of complex systems by parts.
arXiv Detail & Related papers (2022-11-03T15:04:27Z) - Detailed Balanced Chemical Reaction Networks as Generalized Boltzmann
Machines [0.0]
We show how a biochemical computer can use intrinsic chemical noise to perform complex computations.
We also use our explicit physical model to derive thermodynamic costs of inference.
arXiv Detail & Related papers (2022-05-12T18:59:43Z) - Differentiable Scaffolding Tree for Molecular Optimization [47.447362691543304]
We propose differentiable scaffolding tree (DST) that utilizes a learned knowledge network to convert discrete chemical structures to locally differentiable ones.
Our empirical studies show the gradient-based molecular optimizations are both effective and sample efficient.
arXiv Detail & Related papers (2021-09-22T01:16:22Z) - Quantum Markov Chain Monte Carlo with Digital Dissipative Dynamics on
Quantum Computers [52.77024349608834]
We develop a digital quantum algorithm that simulates interaction with an environment using a small number of ancilla qubits.
We evaluate the algorithm by simulating thermal states of the transverse Ising model.
arXiv Detail & Related papers (2021-03-04T18:21:00Z) - Machine learning of solvent effects on molecular spectra and reactions [3.4376560669160394]
We introduce the deep neural network FieldSchNet for modeling the interaction of molecules with arbitrary external fields.
FieldSchNet offers access to a wealth of molecular response properties, enabling it to simulate a wide range of molecular spectra.
We employ FieldSchNet to study the influence of solvent effects on molecular spectra and a Claisen rearrangement reaction.
arXiv Detail & Related papers (2020-10-28T12:50:30Z) - A Quantum Finite Automata Approach to Modeling the Chemical Reactions [0.0]
The study of chemical information processing with quantum computational models is a natural goal.
We have modeled chemical reactions using two-way quantum finite automata, which are halted in linear time.
It has been proven that computational versatility can be increased by combining chemical accept/reject signatures and quantum automata models.
arXiv Detail & Related papers (2020-07-08T09:15:33Z) - Representations of molecules and materials for interpolation of
quantum-mechanical simulations via machine learning [0.0]
Computational study of molecules and materials from first principles is a cornerstone of physics, chemistry, and materials science.
In settings involving many simulations, machine learning can reduce these costs, often by orders of magnitude, by interpolating between reference simulations.
We comprehensively review and discuss current representations and relations between them, using a unified mathematical framework.
arXiv Detail & Related papers (2020-03-26T18:01:35Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.