Search Results for author: Amir Barati Farimani

Found 42 papers, 17 papers with code

Transformer-based Planning for Symbolic Regression

no code implementations13 Mar 2023 Parshin Shojaee, Kazem Meidani, Amir Barati Farimani, Chandan K. Reddy

To address this, we propose TPSR, a Transformer-based Planning strategy for Symbolic Regression that incorporates Monte Carlo Tree Search into the transformer decoding process.

regression Symbolic Regression +1

Denoise Pre-training on Non-equilibrium Molecules for Accurate and Transferable Neural Potentials

1 code implementation3 Mar 2023 Yuyang Wang, Changwen Xu, Zijie Li, Amir Barati Farimani

These results highlight the potential for leveraging denoise pre-training approaches to build more generalizable neural potentials for complex molecular systems.

MeshDQN: A Deep Reinforcement Learning Framework for Improving Meshes in Computational Fluid Dynamics

1 code implementation2 Dec 2022 Cooper Lorsung, Amir Barati Farimani

Current machine learning techniques often require substantial computational cost for training data generation, and are restricted in scope to the training data flow regime.

reinforcement-learning Reinforcement Learning (RL)

A Physics-informed Diffusion Model for High-fidelity Flow Field Reconstruction

1 code implementation26 Nov 2022 Dule Shu, Zijie Li, Amir Barati Farimani

Machine learning models are gaining increasing popularity in the domain of fluid dynamics for their potential to accelerate the production of high-fidelity computational fluid dynamics data.

MOFormer: Self-Supervised Transformer model for Metal-Organic Framework Property Prediction

1 code implementation25 Oct 2022 Zhonglin Cao, Rishikesh Magar, Yuyang Wang, Amir Barati Farimani

Furthermore, we revealed that MOFormer can be more data-efficient on quantum-chemical property prediction than structure-based CGCNN when training data is limited.

Self-Supervised Learning

Predicting CO$_2$ Absorption in Ionic Liquids with Molecular Descriptors and Explainable Graph Neural Networks

1 code implementation29 Sep 2022 Yue Jian, Yuyang Wang, Amir Barati Farimani

Fingerprint works on graph structure at the feature extraction stage, while GNNs directly handle molecule structure in both the feature extraction and model prediction stage.

MAN: Multi-Action Networks Learning

1 code implementation19 Sep 2022 Keqin Wang, Alison Bartsch, Amir Barati Farimani

In this work, we introduce a Deep Reinforcement Learning (DRL) algorithm call Multi-Action Networks (MAN) Learning that addresses the challenge of high-dimensional large discrete action spaces.

Atari Games Q-Learning +2

Graph Neural Networks for Molecules

no code implementations12 Sep 2022 Yuyang Wang, Zijie Li, Amir Barati Farimani

Graph neural networks (GNNs), which are capable of learning representations from graphical data, are naturally suitable for modeling molecular systems.

Molecular Property Prediction Self-Supervised Learning

TransPolymer: a Transformer-based Language Model for Polymer Property Predictions

1 code implementation3 Sep 2022 Changwen Xu, Yuyang Wang, Amir Barati Farimani

The model learns expressive representations by pretraining on a large unlabeled dataset via masked language modeling, followed by finetuning the model on downstream datasets concerning various polymer properties.

Language Modelling Masked Language Modeling +1

Mechanical Properties Prediction in Metal Additive Manufacturing Using Machine Learning

no code implementations21 Aug 2022 Parand Akbari, Ning-Yu Kao, Amir Barati Farimani

Nonetheless, Machine learning (ML) methods, which are more flexible and cost-effective solutions, can be utilized to predict mechanical properties based on the processing parameters and material properties.


Surrogate Modeling of Melt Pool Thermal Field using Deep Learning

no code implementations25 Jul 2022 AmirPouya Hemmasian, Francis Ogoke, Parand Akbari, Jonathan Malen, Jack Beuth, Amir Barati Farimani

In this work, we create three datasets of single-trail processes using Flow-3D and use them to train a convolutional neural network capable of predicting the behavior of the three-dimensional thermal field of the melt pool solely by taking three parameters as input: laser power, laser velocity, and time step.

Transformer for Partial Differential Equations' Operator Learning

no code implementations26 May 2022 Zijie Li, Kazem Meidani, Amir Barati Farimani

Data-driven learning of partial differential equations' solution operators has recently emerged as a promising paradigm for approximating the underlying solutions.

Operator learning

Deep-Learned Generators of Porosity Distributions Produced During Metal Additive Manufacturing

no code implementations11 May 2022 Francis Ogoke, Kyle Johnson, Michael Glinsky, Chris Laursen, Sharlotte Kramer, Amir Barati Farimani

Laser Powder Bed Fusion has become a widely adopted method for metal Additive Manufacturing (AM) due to its ability to mass produce complex parts with increased local control.

Improving Molecular Contrastive Learning via Faulty Negative Mitigation and Decomposed Fragment Contrast

1 code implementation18 Feb 2022 Yuyang Wang, Rishikesh Magar, Chen Liang, Amir Barati Farimani

On most benchmarks, the generic GNN pre-trained by iMolCLR rivals or even surpasses supervised learning models with sophisticated architecture designs and engineered features.

Contrastive Learning Self-Supervised Learning

Graph Neural Networks Accelerated Molecular Dynamics

1 code implementation6 Dec 2021 Zijie Li, Kazem Meidani, Prakarsh Yadav, Amir Barati Farimani

Molecular Dynamics (MD) simulation is a powerful tool for understanding the dynamics and structure of matter.

AugLiChem: Data Augmentation Library of Chemical Structures for Machine Learning

1 code implementation30 Nov 2021 Rishikesh Magar, Yuyang Wang, Cooper Lorsung, Chen Liang, Hariharan Ramasubramanian, Peiyuan Li, Amir Barati Farimani

Inspired by the success of data augmentations in computer vision and natural language processing, we developed AugLiChem: the data augmentation library for chemical structures.

BIG-bench Machine Learning Data Augmentation

TPU-GAN: Learning temporal coherence from dynamic point cloud sequences

1 code implementation ICLR 2022 Zijie Li, Tianqin Li, Amir Barati Farimani

Our model, Temporal Point cloud Upsampling GAN (TPU-GAN), can implicitly learn the underlying temporal coherence from point cloud sequence, which in turn guides the generator to produce temporally coherent output.


Prototype memory and attention mechanisms for few shot image generation

no code implementations ICLR 2022 Tianqin Li, Zijie Li, Andrew Luo, Harold Rockwell, Amir Barati Farimani, Tai Sing Lee

To test our proposal, we show in a few-shot image generation task, that having a prototype memory during attention can improve image synthesis quality, learn interpretable visual concept clusters, as well as improve the robustness of the model.

Image Generation Online Clustering

Deep Learning for Reduced Order Modelling and Efficient Temporal Evolution of Fluid Simulations

1 code implementation9 Jul 2021 Pranshu Pant, Ruchit Doshi, Pranav Bahl, Amir Barati Farimani

Reduced Order Modelling (ROM) has been widely used to create lower order, computationally inexpensive representations of higher-order dynamical systems.

Dimensionality Reduction

Dominant motion identification of multi-particle system using deep learning from video

1 code implementation26 Apr 2021 Yayati Jadhav, Amir Barati Farimani

In this work, we provide a deep-learning framework that extracts relevant information from real-world videos of highly stochastic systems, with no prior knowledge and distills the underlying governing equation representing the system.

Feature Engineering

An Energy-Saving Snake Locomotion Gait Policy Obtained Using Deep Reinforcement Learning

no code implementations8 Mar 2021 Yilang Liu, Amir Barati Farimani

Comparing to conventional control strategies, the snake robots controlled by the trained PPO agent can achieve faster movement and more energy-efficient locomotion gait.

Navigate reinforcement-learning +1

Molecular Contrastive Learning of Representations via Graph Neural Networks

1 code implementation19 Feb 2021 Yuyang Wang, Jianren Wang, Zhonglin Cao, Amir Barati Farimani

In this work, we present MolCLR: Molecular Contrastive Learning of Representations via Graph Neural Networks (GNNs), a self-supervised learning framework that leverages large unlabeled data (~10M unique molecules).

BIG-bench Machine Learning Contrastive Learning +3

Thermal Control of Laser Powder Bed Fusion Using Deep Reinforcement Learning

no code implementations29 Jan 2021 Francis Ogoke, Amir Barati Farimani

Powder-based additive manufacturing techniques provide tools to construct intricate structures that are difficult to manufacture using conventional methods.

reinforcement-learning Reinforcement Learning (RL)

Deep Reinforcement Learning Optimizes Graphene Nanopores for Efficient Desalination

no code implementations19 Jan 2021 Yuyang Wang, Zhonglin Cao, Amir Barati Farimani

Structure and geometry optimization of nanopores on such materials is beneficial for their performances in real-world engineering applications, like water desalination.

reinforcement-learning Reinforcement Learning (RL)

Airfoil GAN: Encoding and Synthesizing Airfoils forAerodynamic-aware Shape Optimization

no code implementations12 Jan 2021 Yuyang Wang, Kenji Shimada, Amir Barati Farimani

Our model can (1) encode the existing airfoil into a latent vector and reconstruct the airfoil from that, (2) generate novel airfoils by randomly sampling the latent vectors and mapping the vectors to the airfoil coordinate domain, and (3) synthesize airfoils with desired aerodynamic properties by optimizing learned features via a genetic algorithm.

Learning Lagrangian Fluid Dynamics with Graph Neural Networks

no code implementations1 Jan 2021 Zijie Li, Amir Barati Farimani

We present a data-driven model for fluid simulation under Lagrangian representation.

Graph Convolutional Neural Networks for Body Force Prediction

no code implementations3 Dec 2020 Francis Ogoke, Kazem Meidani, Amirreza Hashemi, Amir Barati Farimani

The ability of the method to predict global properties from spatially irregular measurements with high accuracy is demonstrated by predicting the drag force associated with laminar flow around airfoils from scattered velocity measurements.

Deep Learning for Efficient Reconstruction of High-Resolution Turbulent DNS Data

1 code implementation21 Oct 2020 Pranshu Pant, Amir Barati Farimani

Large Eddy Simulation (LES) presents a more computationally efficient approach for solving fluid flows on lower-resolution (LR) grids but results in an overall reduction in solution fidelity.

Image Super-Resolution

FaultNet: A Deep Convolutional Neural Network for bearing fault classification

1 code implementation5 Oct 2020 Rishikesh Magar, Lalit Ghule, Junhan Li, Yang Zhao, Amir Barati Farimani

In this work, we analyze vibration signal data of mechanical systems with bearings by combining different signal processing methods and coupling them with machine learning techniques to classify different types of bearing faults.

BIG-bench Machine Learning Classification +2

Orbital Graph Convolutional Neural Network for Material Property Prediction

no code implementations14 Aug 2020 Mohammadreza Karamad, Rishikesh Magar, Yuting Shi, Samira Siahrostami, Ian D. Gates, Amir Barati Farimani

We benchmarked the performance of the OGCNN model with that of: 1) the crystal graph convolutional neural network (CGCNN), 2) other state-of-the-art descriptors for material representations including Many-body Tensor Representation (MBTR) and the Smooth Overlap of Atomic Positions (SOAP), and 3) other conventional regression machine learning algorithms where different crystal featurization methods have been used.

BIG-bench Machine Learning

Deep Learning Convective Flow Using Conditional Generative Adversarial Networks

1 code implementation13 May 2020 Changlin Jiang, Amir Barati Farimani

We developed a general deep learning framework, FluidGAN, that is capable of learning and predicting time-dependent convective flow coupled with energy transport.

Potential Neutralizing Antibodies Discovered for Novel Corona Virus Using Machine Learning

no code implementations18 Mar 2020 Rishikesh Magar, Prakarsh Yadav, Amir Barati Farimani

The fast and untraceable virus mutations take lives of thousands of people before the immune system can produce the inhibitory antibody.

BIG-bench Machine Learning

Effects of sparse rewards of different magnitudes in the speed of learning of model-based actor critic methods

no code implementations18 Jan 2020 Juan Vargas, Lazar Andjelic, Amir Barati Farimani

Actor critic methods with sparse rewards in model-based deep reinforcement learning typically require a deterministic binary reward function that reflects only two possible outcomes: if, for each step, the goal has been achieved or not.

Weakly-Supervised Deep Learning of Heat Transport via Physics Informed Loss

no code implementations24 Jul 2018 Rishi Sharma, Amir Barati Farimani, Joe Gomes, Peter Eastman, Vijay Pande

In typical machine learning tasks and applications, it is necessary to obtain or create large labeled datasets in order to to achieve high performance.

BIG-bench Machine Learning

Deep Learning Phase Segregation

no code implementations23 Mar 2018 Amir Barati Farimani, Joseph Gomes, Rishi Sharma, Franklin L. Lee, Vijay S. Pande

Phase segregation, the process by which the components of a binary mixture spontaneously separate, is a key process in the evolution and design of many chemical, mechanical, and biological systems.

Machine Learning Harnesses Molecular Dynamics to Discover New $μ$ Opioid Chemotypes

no code implementations12 Mar 2018 Evan N. Feinberg, Amir Barati Farimani, Rajendra Uprety, Amanda Hunkele, Gavril W. Pasternak, Susruta Majumdar, Vijay S. Pande

Computational chemists typically assay drug candidates by virtually screening compounds against crystal structures of a protein despite the fact that some targets, like the $\mu$ Opioid Receptor and other members of the GPCR family, traverse many non-crystallographic states.

BIG-bench Machine Learning

Deep Learning the Physics of Transport Phenomena

no code implementations7 Sep 2017 Amir Barati Farimani, Joseph Gomes, Vijay S. Pande

We have developed a new data-driven paradigm for the rapid inference, modeling and simulation of the physics of transport phenomena by deep learning.

Cannot find the paper you are looking for? You can Submit a new open access paper.