Corpus ID: 226278128

Is Transfer Learning Necessary for Protein Landscape Prediction?

@article{Shanehsazzadeh2020IsTL,
  title={Is Transfer Learning Necessary for Protein Landscape Prediction?},
  author={Amir Shanehsazzadeh and David Belanger and David Dohan},
  journal={ArXiv},
  year={2020},
  volume={abs/2011.03443}
}
Recently, there has been great interest in learning how to best represent proteins, specifically with fixed-length embeddings. Deep learning has become a popular tool for protein representation learning as a model's hidden layers produce potentially useful vector embeddings. TAPE introduced a number of benchmark tasks and showed that semi-supervised learning, via pretraining language models on a large protein corpus, improved performance on downstream tasks. Two of the tasks (fluorescence… Expand
2 Citations

Figures and Tables from this paper

Single Layers of Attention Suffice to Predict Protein Contacts
  • 2
  • PDF
Protein sequence design with deep generative models
  • PDF

References

SHOWING 1-10 OF 19 REFERENCES
Evaluating Protein Transfer Learning with TAPE
  • 88
  • PDF
Learned protein embeddings for machine learning
  • 75
  • PDF
Learning protein sequence embeddings using information from structure
  • 60
  • PDF
Adam: A Method for Stochastic Optimization
  • 60,950
  • PDF
...
1
2
...