Search results with tag "Neural information processing systems"
A Unified Approach to Interpreting Model Predictions - NIPS
papers.nips.cc31st Conference on Neural Information Processing Systems (NIPS 2017), Long Beach, CA, USA. 2. We then show that game theory results guaranteeing a unique solution apply to the entire class of additive feature attribution methods (Section 3) and propose SHAP values as …
Attention is All you Need - NIPS
papers.nips.cc31st Conference on Neural Information Processing Systems (NIPS 2017), Long Beach, CA, USA. Recurrent models typically factor computation along the symbol positions of the input and output sequences. Aligning the positions to steps in computation time, they generate a …
Supervised Contrastive Learning - NIPS
papers.nips.cc34th Conference on Neural Information Processing Systems (NeurIPS 2020), Vancouver, Canada. Figure 2: Supervised vs. self-supervised contrastive losses: The self-supervised contrastive loss (left, Eq.1) contrasts a single positive for each anchor (i.e., an augmented version of the same image) against a set of
![arXiv:1706.02216v4 [cs.SI] 10 Sep 2018](/cache/preview/9/a/6/8/0/7/1/d/thumb-9a68071d73afd8997ff3a74956baa9eb.jpg)
arXiv:1706.02216v4 [cs.SI] 10 Sep 2018
arxiv.org31st Conference on Neural Information Processing Systems (NIPS 2017), Long Beach, CA, USA. arXiv:1706.02216v4 [cs.SI] 10 Sep 2018. Figure 1: Visual illustration of the GraphSAGE sample and aggregate approach. recognize structural properties of a node’s neighborhood that reveal both the node’s local role in the
Neural Discrete Representation Learning
arxiv.org31st Conference on Neural Information Processing Systems (NIPS 2017), Long Beach, CA, USA. arXiv:1711.00937v2 [cs.LG] 30 May 2018 “posterior collapse” issue which has been problematic with many VAE models that have a powerful decoder, often caused by latents being ignored. Additionally, it is the first discrete latent VAE model