Display Week Home Page

Session and scheduling information are listed below. Select a session from the list and press "Go" to view the abstracts for that session.

Session 85 :
AI/ML for OLEDs
OLEDs ; Artificial Intelligence Including Machine Learning for Imaging

Friday, May 17 / 09:00 AM   - 10:20 AM / San Jose Convention Center, LL21CD

Chair:
Eunkyung Koh, Samsung Display Research Center, Yongin, South Korea

Co-Chair:
Yifan Zhang, Apple, Inc., Cupertino, CA US

85.1 - Invited Paper: A Novel OLED Material Discovery Based on AI Technology (9:00 AM - 9:20 AM)
  • Hoilim Kim, Seran Kim, Dongsun Yoo, Gyeonghun Kim, Eunkyung Koh, Jihye Kim, Saerom Park, Sohae Kim, Hyosup Shin, Hyunguk Cho, Seungin Baek
    Samsung Display Gyeonggi-do South Korea


  • A novel OLED material discovery process based on AI technology is reported. Six AI modules generate molecular structures using an active learning algorithm and predict multiple properties, novelty, synthetic scheme, relative synthesizability, and device characteristics. Some evaluation results are introduced to confirm the improvement.
05/17/2024 9:00 AM 05/17/2024 9:20 AM America/Los_Angeles A Novel OLED Material Discovery Based on AI Technology A novel OLED material discovery process based on AI technology is reported. Six AI modules generate molecular structures using an active learning algorithm and predict multiple properties, novelty, synthetic scheme, relative synthesizability, and device characteristics. Some evaluation results are introduced to confirm the improvement. San Jose McEnery Convention Center LL21CD Hoilim Kim, Seran Kim, Dongsun Yoo, Gyeonghun Kim, Eunkyung Koh, Jihye Kim, Saerom Park, Sohae Kim, Hyosup Shin, Hyunguk Cho, Seungin Baek
85.2 - Prediction of Triplet Harvesting Ability in Blue Fluorescent Organic Light-Emitting Diodes Using Deep Learning (9:20 AM - 9:40 AM)
  • Junseop Lim, Jun Yeob Lee
    Sungkyunkwan University Suwon South Korea

  • Jae-Min Kim
    Chung-Ang University Anseong South Korea


  • The authors implemented a new decay model based on exciton dynamics using transient electroluminescence of fluorescent triplet-triplet annihilation OLEDs. In addition, a predictive model was established using a neural network of multilayer perception, and it demonstrated nearly perfect prediction ability of TTA ratio (determination coefficient, R2 = 0.999). 
05/17/2024 9:20 AM 05/17/2024 9:40 AM America/Los_Angeles Prediction of Triplet Harvesting Ability in Blue Fluorescent Organic Light-Emitting Diodes Using Deep Learning The authors implemented a new decay model based on exciton dynamics using transient electroluminescence of fluorescent triplet-triplet annihilation OLEDs. In addition, a predictive model was established using a neural network of multilayer perception, and it demonstrated nearly perfect prediction ability of TTA ratio (determination coefficient, R2 = 0.999).  San Jose McEnery Convention Center LL21CD Jae-Min Kim
85.3 - Machine Learning Strategy Toward Inverse Design of Blue TADF Emitter: Training Excited State Properties Based on Density Functional Theory Calculations (9:40 AM - 10:00 AM)
  • Hyun-Jung Kim, Junho Lee, Yeol Kyo Choi, Taeyang Lee, Joong-Hwan Yang, Jeongguk Min, Ji-Ho Baek, Seok-Woo Lee, Joon-Young Yang, Soo-Young Yoon
    LG Display Co., Ltd. Seoul South Korea

  • Sung Moon Ko, Dae-Woong Jeong, Sehui Han
    LG AI Research Seoul South Korea


  • The authors introduce inverse design strategy utilizing machine learning to discover efficient blue thermally activated delayed fluorescence organic emitters. They utilize a graph neural network to predict the characteristic intrinsic material properties. They discuss consistency between experimental observation and predictions, and examine conditions for improving the accuracy of density-functional theory calculations.
05/17/2024 9:40 AM 05/17/2024 10:00 AM America/Los_Angeles Machine Learning Strategy Toward Inverse Design of Blue TADF Emitter: Training Excited State Properties Based on Density Functional Theory Calculations The authors introduce inverse design strategy utilizing machine learning to discover efficient blue thermally activated delayed fluorescence organic emitters. They utilize a graph neural network to predict the characteristic intrinsic material properties. They discuss consistency between experimental observation and predictions, and examine conditions for improving the accuracy of density-functional theory calculations. San Jose McEnery Convention Center LL21CD Sung Moon Ko, Dae-Woong Jeong, Sehui Han
85.4 - Digital Chemistry, Data Processing, and Collaborative Ideation for Development of OLEDs (10:00 AM - 10:20 AM)
  • Hadi Abroshan, H. Shaun Kwak, Anand Chandrasekaran, Eric Collins, Paul Winget, David Geisen, Thomas Mustard, Yun Liu, Christopher Brown, Matthew Halls
    Schrödinger Inc. New York NY US


  • Empowered by digital chemistry and informatics platforms, this study highlights the transformative impact of physics-based simulation, machine learning, and data management in display industry R&D. This technology integration accelerates ideation and decision-making, ensuring swift, accurate, and cost-effective development for next-generation display devices.
05/17/2024 10:00 AM 05/17/2024 10:20 AM America/Los_Angeles Digital Chemistry, Data Processing, and Collaborative Ideation for Development of OLEDs Empowered by digital chemistry and informatics platforms, this study highlights the transformative impact of physics-based simulation, machine learning, and data management in display industry R&D. This technology integration accelerates ideation and decision-making, ensuring swift, accurate, and cost-effective development for next-generation display devices. San Jose McEnery Convention Center LL21CD Hadi Abroshan, H. Shaun Kwak, Anand Chandrasekaran, Eric Collins, Paul Winget, David Geisen, Thomas Mustard, Yun Liu, Christopher Brown, Matthew Halls