It only uses the learnable prompt vector and not the GNN weights, which are frozen for downstream tasks. This reduces the number of parameters that need to be updated, improves the computational efficiency of task learning/inference, and reduces the reliance on labeled data.
@maximelabonne
-
Prompt Tuning Optimizes Downstream Task Efficiency and Accuracy
By
–
Prompt tuning involves optimizing the learnable prompt to improve the computational efficiency and accuracy of downstream tasks. It is based on the similarity of subgraphs and is formulated using prompt-assisted task-specific subgraph representations.
-
Learnable Prompts Enable Better Task-Specific Knowledge Extraction
By
–
Learnable prompts provide a better alternative to handcrafted prompts, and enable the extraction of the most relevant prior knowledge for each task. The prompts are a dimension-wise reweighting (or a linear transformation) of the node representations.
-
Graph Neural Networks Pre-training with Link Prediction Task
By
–
During pre-training, the GNN is trained on a link prediction task. By sampling nodes and forming triplets, a pre-training loss is constructed that improves the similarity between the contextual subgraphs of two candidate nodes.
-
GraphPrompt: Unified Pre-Training Framework for Graph Neural Networks
By
–
GraphPrompt: Unifying Pre-Training and Downstream Tasks for Graph Neural Networks Interesting pre-training training framework for GNNs, based on learnable prompts. It reduces labeling needs and boosts downstream task performance. https://
arxiv.org/abs/2302.08043 -
Graph Neural Networks for Optimization and Scheduling Solutions
By
–
Not directly: I'm interested in this field in general because there are a lot of cool applications, like optimizing schedules. But yes, I'm also interested in building a GNN that acts as a solver or just helps with optimization in general, like https://
arxiv.org/pdf/2209.12288
.pdf
… -
Convex Optimization Comprehensive Resource PDF Available
By
–
Received this morning! Convex Optimization is widely regarded as the most comprehensive resource on convex optimization. If you're interested, here's the full pdf version: https://
web.stanford.edu/~boyd/cvxbook/ -
GPT-Based Synthetic Tabular Data Generation Library
By
–
Cool new #GPT-based approach to synthetic tabular data generation. It shows great results compared to the excellent SDV library. Library: https://
github.com/avsolatorio/re
altabformer
… -
ML Professional Launches GNN Book, Joins JPMorgan UK Team
By
–
Back on Twitter after a few chaotic months! A few updates: – I moved to the UK and joined an excellent machine learning team @jpmorgan – I'm publishing a book about Graph Neural Networks for practitioners Stay tuned for more 🙂