Measuring Compositional Generalization: A Comprehensive Method on Realistic Data
December 20, 2019 Β· Declared Dead Β· π International Conference on Learning Representations
"No code URL or promise found in abstract"
Evidence collected by the PWNC Scanner
Authors
Daniel Keysers, Nathanael SchΓ€rli, Nathan Scales, Hylke Buisman, Daniel Furrer, Sergii Kashubin, Nikola Momchev, Danila Sinopalnikov, Lukasz Stafiniak, Tibor Tihon, Dmitry Tsarkov, Xiao Wang, Marc van Zee, Olivier Bousquet
arXiv ID
1912.09713
Category
cs.LG: Machine Learning
Cross-listed
cs.CL,
stat.ML
Citations
389
Venue
International Conference on Learning Representations
Last Checked
3 months ago
Abstract
State-of-the-art machine learning methods exhibit limited compositional generalization. At the same time, there is a lack of realistic benchmarks that comprehensively measure this ability, which makes it challenging to find and evaluate improvements. We introduce a novel method to systematically construct such benchmarks by maximizing compound divergence while guaranteeing a small atom divergence between train and test sets, and we quantitatively compare this method to other approaches for creating compositional generalization benchmarks. We present a large and realistic natural language question answering dataset that is constructed according to this method, and we use it to analyze the compositional generalization ability of three machine learning architectures. We find that they fail to generalize compositionally and that there is a surprisingly strong negative correlation between compound divergence and accuracy. We also demonstrate how our method can be used to create new compositionality benchmarks on top of the existing SCAN dataset, which confirms these findings.
Community Contributions
Found the code? Know the venue? Think something is wrong? Let us know!
π Similar Papers
In the same crypt β Machine Learning
R.I.P.
π»
Ghosted
R.I.P.
π»
Ghosted
XGBoost: A Scalable Tree Boosting System
R.I.P.
π»
Ghosted
Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
R.I.P.
π»
Ghosted
Semi-Supervised Classification with Graph Convolutional Networks
R.I.P.
π»
Ghosted
Proximal Policy Optimization Algorithms
R.I.P.
π»
Ghosted
Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer
Died the same way β π» Ghosted
R.I.P.
π»
Ghosted
Language Models are Few-Shot Learners
R.I.P.
π»
Ghosted
You Only Look Once: Unified, Real-Time Object Detection
R.I.P.
π»
Ghosted
A Unified Approach to Interpreting Model Predictions
R.I.P.
π»
Ghosted