Multi-View Mixture-of-Experts for Predicting Molecular Properties Using SMILES, SELFIES, and Graph-Based Representations
- 2024
- NeurIPS 2024
Bio
Indra Priyadarsini joined IBM Research - Tokyo in January 2023.
She recieved her B.E in Electronics and Communication Engineering from PES Institute of Technology, Bangalore, India in 2016, and her M.E and Ph.D degree from Shizuoka University, Japan in 2019 and 2022, respectively.
Her research interests include optimization algorithms and deep learning.
Publications
SELF-BART : A Transformer-based Molecular Representation Model using SELFIES
- Indra Priyadarsini S
- Seiji Takeda
- et al.
- 2024
- NeurIPS 2024
A Dynamic Multi-Modal Fusion Model for Material Discovery
- Indra Priyadarsini S
- Seiji Takeda
- et al.
- 2024
- MRS Fall Meeting 2024
SELF-BART : A Transformer-based Molecular Representation Model using SELFIES
- Indra Priyadarsini S
- Seiji Takeda
- et al.
- 2024
- arXiv
Improving Performance Prediction of Electrolyte Formulations with Transformer-based Molecular Representation Model
- 2024
- ICML 2024
Improving Performance Prediction of Electrolyte Formulations with Transformer-based Molecular Representation Model
- 2024
- arXiv
MHG-GNN: Combination of Molecular Hypergraph Grammar with Graph Neural Network
- Akihiro Kishimoto
- Hiroshi Kajino
- et al.
- 2023
- NeurIPS 2023
Pre-Treatment Methods for Machine Learning in Finer UV Spectrum Inference
- 2023
- MRS Fall Meeting 2023
Multi-Modal Foundation Model for Material Discovery with Conversational User Interface
- Seiji Takeda
- Akihiro Kishimoto
- et al.
- 2023
- MRS Fall Meeting 2023
A transformer based large-scale molecular representation model
- Indra Priyadarsini S
- Seiji Takeda
- et al.
- 2023
- MRS Fall Meeting 2023
Top collaborators
HS
Hajime Shinohara
TA of Research - Tokyo / Research Scientist / Happy Friday Staff
LH
Lisa Hamada
Research Scientist
DN
Daiju Nakano
Technical Lead of Quantum Hardware in Japan
EA
Eduardo Soares
Research Scientist