Yunchao (Lance) Liu 刘运超

Published in Towards Data Science

·Pinned

MolKGNN: Extending Convolution To Molecules

Understand MolKGNN, an interpretable GNN tailored for drug discovery — This blog introduces our latest model Molecular Kernel Graph Neural Network (MolKGNN) from the paper Interpretable Chirality-Aware Graph Neural Network for Quantitative Structure Activity Relationship Modeling in Drug Discovery Introduction: Elucidating molecular structures and their pharmacological activity has been a long-standing problem in the history of drug discovery. In 1859, the…

Drug Discovery

9 min read

MolKGNN: Extending Convolution To Molecules

Drug Discovery

9 min read

Published in MLearning.ai

·Pinned

Read Paper with Me: Data2Vec: A General Framework for Self-supervised Learning in Speech, Vision and Language

Interpretation of data2vec, a model for learning across modalities — This blog introduces a new paper on self-supervised learning from Meta AI: data2vec: A General Framework for Self-supervised Learning in Speech, Vision, and Language If you have a hard time understanding this blog, it’s recommended to read my blog on BYOL first. This work is built upon BYOL. Background: Traditional…

Machine Learning

6 min read

Meta AI Published Data2Vec. A New Milestone for Learning Across Modalities

Machine Learning

6 min read

Published in MLearning.ai

·Jan 29, 2022

Read Papers With Me: Bootstrap Your Own Latent: A New Approach To Self-Supervised Learning

Interpretation of the BYOL, a self-supervised learning method — This blog introduces the Bootstrap Your Own Latent (BYOL) method, a self-supervised learning method proposed by DeepMind and Imperial London A glance at the paper: BYOL contains two networks with the same architecture, but different parameters. BYOL does NOT need negative pairs, which most contrastive learning methods do, such as…

Data Science

4 min read

Read Paper With Me: Bootstrap Your Own Latent: A New Approach To Self-Supervised Learning

Data Science

4 min read