Skip to content

Latest commit

 

History

History

research

Research projects in Neural Structured Learning

Note that these research projects are not included in the prebuilt NSL pip package.

The implementations of Low-Dimensional Hyperbolic Knowledge Graph Embeddings [3] are provided in the kg_hyp_emb folder on a strict "as is" basis, without warranties or conditions of any kind. Also, these implementations may not be compatible with certain TensorFlow versions or Python versions.

[3] Chami, Ines, et al. "Low-Dimensional Hyperbolic Knowledge Graph Embeddings." ACL 2020.

A2N: Attending to Neighbors for Knowledge Graph Inference

The implementations of A2N [2] are provided in the a2n folder on a strict "as is" basis, without warranties or conditions of any kind. Also, these implementations may not be compatible with certain TensorFlow versions or Python versions.

[2] T. Bansal, D. Juan, S. Ravi and A. McCallum. "A2N: Attending to Neighbors for Knowledge Graph Inference." ACL 2019

GAM: Graph Agreement Models for Semi-Supervised Learning

The implementations of Graph Agreement Models (GAMs) [1] are provided in the gam folder on a strict "as is" basis, without warranties or conditions of any kind. Also, these implementations may not be compatible with certain TensorFlow versions or Python versions.

[1] O. Stretcu, K. Viswanathan, D. Movshovitz-Attias, E.A. Platanios, S. Ravi, A. Tomkins. "Graph Agreement Models for Semi-Supervised Learning." NeurIPS 2019

The implementations of Neural Clustering Process (NCP) [4] are provided in the neural_clustering folder on a strict "as is" basis, without warranties or conditions of any kind. Also, these implementations may not be compatible with certain TensorFlow versions or Python versions.

[4] A. Pakman, Y. Wang, C. Mitelut, J. Lee, L. Paninski. "Neural Clustering Processes." ICML 2020

The implementations of Denoised Smoothing [5] are provided in the third_party/denoised_smoothing folder on a strict "as is" basis, without warranties or conditions of any kind. Also, these implementations may not be compatible with certain TensorFlow versions or Python versions.

[5] H. Salman, M.Sun, C. Mitelut, G. Yang, A. Kapoor., Z. Kolter "Denoised Smoothing: A Provable Defense for Pretrained Classifiers." NeurIPS 2020