Hardik Rajpal
1 indexed paper
Recent (6 mo)
1With code
0Influential cites
0Benchmarked
0Publications per year
126
Top categories
Neurons and Cognition×1Info Theory×1ML×1Neural Computing×1nlin.CD×1
Frequent co-authors
Research Timeline
2026
Emergent Generalization by Representation Learning in Artificial Neural Networks
The authors show that an explicit information bottleneck in a recurrent neural network is necessary for rotational and out-of-distribution generalization in a time-series prediction task, and that the emergence of structure in the representation scales with task complexity and predicts generalization performance.
Highlighted terms show continued research focus across papers