mlsanityData to 7 Oct 2026

Multi-Objective Latent Space Optimization of Generative Molecular Design Models

arXiv 2203.00526 ↗First posted 1 Mar 2022cs.LG q-bio.BMcited by 13

Not on the map. Why?

Abstract (arXiv)

Molecular design based on generative models, such as variational autoencoders (VAEs), has become increasingly popular in recent years due to its efficiency for exploring high-dimensional molecular space to identify molecules with desired properties. While the efficacy of the initial model strongly depends on the training data, the sampling efficiency of the model for suggesting novel molecules with enhanced properties can be further enhanced via latent space optimization. In this paper, we propose a multi-objective latent space optimization (LSO) method that can significantly enhance the performance of generative molecular design (GMD). The proposed method adopts an iterative weighted retraining approach, where the respective weights of the molecules in the training data are determined by their Pareto efficiency. We demonstrate that our multi-objective GMD LSO method can significantly improve the performance of GMD for jointly optimizing multiple molecular properties.

Key facts. Cited by 13 papers in the map (data through 7 Oct 2026). Builds on 16 of its 57 references that are in the map. First posted on arXiv 1 Mar 2022.

Builds on

16 of 57 references.

References · 16

  1. 2016Automatic chemical design using a data-driven continuous representation of moleculescited by 762
  2. 2018A Tutorial on Bayesian Optimizationcited by 734
  3. 2018Junction Tree Variational Autoencoder for Molecular Graph Generationcited by 647
  4. 2020GraphAF: a Flow-based Autoregressive Model for Molecular Graph Generationcited by 301
  5. 2017Grammar Variational Autoencodercited by 283
  6. 2017Molecular De Novo Design through Deep Reinforcement Learningcited by 264
  7. 2018Optimization of Molecules via Deep Reinforcement Learningcited by 199
  8. 2021MARS: Markov Molecular Sampling for Multi-objective Drug Discoverycited by 125
  9. 2020Sample-Efficient Optimization in the Latent Space of Deep Generative Models via Weighted Retrainingcited by 121
  10. 2021Improving black-box optimization in VAE latent space using decoder uncertaintycited by 51
  11. 2018Conditional molecular design with deep generative modelscited by 44
  12. 2020Improved Protein-ligand Binding Affinity Prediction with Structure-Based Deep Fusion Inferencecited by 35
4 more
  1. 2020Improving Molecular Design by Stochastic Iterative Target Augmentationcited by 18
  2. 2020Quantifying the multi-objective cost of uncertaintycited by 5
  3. 2020Optimal Experimental Design for Uncertain Systems Based on Coupled Differential Equationscited by 4
  4. 2021Accelerating Optimal Experimental Design for Robust Synchronization of Uncertain Kuramoto Oscillator Model Using Machine Learningcited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 1; 2023: 2; 2024: 3; 2025: 2; 2026: 5.

13 papers.

Sort:

By area:

  1. 2022Computer-Aided Multi-Objective Optimization in Small Molecule Discoverycited by 31
  2. 2023Advancing Bayesian Optimization via Learning Correlated Latent Spacecited by 18
  3. 2023ChemSpaceAL: An Efficient Active Learning Methodology Applied to Protein-Specific Molecular Generationcited by 8
  4. 2025ExLLM: Experience-Enhanced LLM Optimization for Molecular Design and Beyondcited by 6
  5. 2024Leveraging Active Subspaces to Capture Epistemic Model Uncertainty in Deep Generative Models for Molecular Designcited by 2
  6. 2024Enhancing Generative Molecular Design via Uncertainty-guided Fine-tuning of Variational Autoencoderscited by 2
  7. 2025FragmentGPT: A Unified GPT Model for Fragment Growing, Linking, and Merging in Molecular Designcited by 1
  8. 2026MEIDNet: Multimodal generative AI framework for inverse materials designcited by 1
  9. 2026mRNAutilus: Multi-Objective-Guided Discrete Generation of mRNA with Optimized Therapeutic Propertiescited by 1
  10. 2024Pathway-Guided Optimization of Deep Generative Molecular Design Models for Cancer Therapycited by 0
  11. 2026Q-BIOLAT: Binary Latent Protein Fitness Landscapes for QUBO-Based Optimizationcited by 0
  12. 2026Curated Synthetic Data Doesn't Have to Collapse: A Theoretical Study of Generative Retraining with Pluralistic Preferencescited by 0
1 more
  1. 2026Don't Retrain, Just Reuse: Recovering Dual-Target Molecules from Single-Target Diffusion Modelscited by 0