3D Denoisers are Good 2D Teachers: Molecular Pretraining via Denoising and Cross-Modal Distillation

• URL: http://arxiv.org/abs/2309.04062v1
• Date: Fri, 8 Sep 2023 01:36:58 GMT
• Title: 3D Denoisers are Good 2D Teachers: Molecular Pretraining via Denoising and Cross-Modal Distillation
• Authors: Sungjun Cho, Dae-Woong Jeong, Sung Moon Ko, Jinwoo Kim, Sehui Han, Seunghoon Hong, Honglak Lee, Moontae Lee
• Abstract summary: We propose D&D, a self-supervised molecular representation learning framework that pretrains a 2D graph encoder by distilling representations from a 3D denoiser. We show that D&D can infer 3D information based on the 2D graph and shows superior performance and label-efficiency against other baselines.
• Score: 65.35632020653291
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Pretraining molecular representations from large unlabeled data is essential for molecular property prediction due to the high cost of obtaining ground-truth labels. While there exist various 2D graph-based molecular pretraining approaches, these methods struggle to show statistically significant gains in predictive performance. Recent work have thus instead proposed 3D conformer-based pretraining under the task of denoising, which led to promising results. During downstream finetuning, however, models trained with 3D conformers require accurate atom-coordinates of previously unseen molecules, which are computationally expensive to acquire at scale. In light of this limitation, we propose D&D, a self-supervised molecular representation learning framework that pretrains a 2D graph encoder by distilling representations from a 3D denoiser. With denoising followed by cross-modal knowledge distillation, our approach enjoys use of knowledge obtained from denoising as well as painless application to downstream tasks with no access to accurate conformers. Experiments on real-world molecular property prediction datasets show that the graph encoder trained via D&D can infer 3D information based on the 2D graph and shows superior performance and label-efficiency against other baselines.

Related papers

• 3D-Mol: A Novel Contrastive Learning Framework for Molecular Property Prediction with 3D Information [1.1777304970289215]
  3D-Mol is a novel approach designed for more accurate spatial structure representation. It deconstructs molecules into three hierarchical graphs to better extract geometric information. We compare 3D-Mol with various state-of-the-art baselines on 7 benchmarks and demonstrate our outstanding performance.
  arXiv  Detail & Related papers  (2023-09-28T10:05:37Z)
• Geometry-aware Line Graph Transformer Pre-training for Molecular Property Prediction [4.598522704308923]
  Geometry-aware line graph transformer (Galformer) pre-training is a novel self-supervised learning framework. Galformer consistently outperforms all baselines on both classification and regression tasks.
  arXiv  Detail & Related papers  (2023-09-01T14:20:48Z)
• Automated 3D Pre-Training for Molecular Property Prediction [54.15788181794094]
  We propose a novel 3D pre-training framework (dubbed 3D PGT) It pre-trains a model on 3D molecular graphs, and then fine-tunes it on molecular graphs without 3D structures. Extensive experiments on 2D molecular graphs are conducted to demonstrate the accuracy, efficiency and generalization ability of the proposed 3D PGT.
  arXiv  Detail & Related papers  (2023-06-13T14:43:13Z)
• Video Pretraining Advances 3D Deep Learning on Chest CT Tasks [63.879848037679224]
  Pretraining on large natural image classification datasets has aided model development on data-scarce 2D medical tasks. These 2D models have been surpassed by 3D models on 3D computer vision benchmarks. We show video pretraining for 3D models can enable higher performance on smaller datasets for 3D medical tasks.
  arXiv  Detail & Related papers  (2023-04-02T14:46:58Z)
• 3D Point Cloud Pre-training with Knowledge Distillation from 2D Images [128.40422211090078]
  We propose a knowledge distillation method for 3D point cloud pre-trained models to acquire knowledge directly from the 2D representation learning model. Specifically, we introduce a cross-attention mechanism to extract concept features from 3D point cloud and compare them with the semantic information from 2D images. In this scheme, the point cloud pre-trained models learn directly from rich information contained in 2D teacher models.
  arXiv  Detail & Related papers  (2022-12-17T23:21:04Z)
• 3D Equivariant Molecular Graph Pretraining [42.957880677779556]
  We tackle 3D molecular pretraining in a complete and novel sense. We first propose to adopt an equivariant energy-based model as the backbone for pretraining, which enjoys the merit of fulfilling the symmetry of 3D space. We evaluate our model pretrained from a large-scale 3D dataset GEOM-QM9 on two challenging 3D benchmarks: MD17 and QM9.
  arXiv  Detail & Related papers  (2022-07-18T16:26:24Z)
• Unified 2D and 3D Pre-Training of Molecular Representations [237.36667670201473]
  We propose a new representation learning method based on a unified 2D and 3D pre-training. Atom coordinates and interatomic distances are encoded and then fused with atomic representations through graph neural networks. Our method achieves state-of-the-art results on 10 tasks, and the average improvement on 2D-only tasks is 8.3%.
  arXiv  Detail & Related papers  (2022-07-14T11:36:56Z)
• Pre-training via Denoising for Molecular Property Prediction [53.409242538744444]
  We describe a pre-training technique that utilizes large datasets of 3D molecular structures at equilibrium. Inspired by recent advances in noise regularization, our pre-training objective is based on denoising.
  arXiv  Detail & Related papers  (2022-05-31T22:28:34Z)
• Advancing 3D Medical Image Analysis with Variable Dimension Transform based Supervised 3D Pre-training [45.90045513731704]
  This paper revisits an innovative yet simple fully-supervised 3D network pre-training framework. With a redesigned 3D network architecture, reformulated natural images are used to address the problem of data scarcity. Comprehensive experiments on four benchmark datasets demonstrate that the proposed pre-trained models can effectively accelerate convergence.
  arXiv  Detail & Related papers  (2022-01-05T03:11:21Z)
• 3D Infomax improves GNNs for Molecular Property Prediction [1.9703625025720701]
  We propose pre-training a model to reason about the geometry of molecules given only their 2D molecular graphs. We show that 3D pre-training provides significant improvements for a wide range of properties.
  arXiv  Detail & Related papers  (2021-10-08T13:30:49Z)

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.