Ruslan Salakhutdinov
Ruslan Salakhutdinov
collaborator
line weight = collaboration weight
Works in this corpus
| Title | Position | Year | Cited |
|---|---|---|---|
| Dropout: a simple way to prevent neural networks from overfitting | last | 2014 | 34,236 |
| Reducing the Dimensionality of Data with Neural Networks | last | 2006 | 21,247 |
| Improving neural networks by preventing co-adaptation of feature detectors | last | 2012 | 6,653 |
| Siamese Neural Networks for One-shot Image Recognition | last | 2015 | 3,253 |
| Transformer-XL: Attentive Language Models beyond a Fixed-Length Context | last | 2019 | 3,202 |
| Human-level concept learning through probabilistic program induction | middle | 2015 | 2,958 |
| XLNet: Generalized Autoregressive Pretraining for Language Understanding | middle | 2019 | 1,854 |
| HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering | middle | 2018 | 1,743 |
Is this one person?
This record may describe more than one person. It is shown as the source publishes it and has not been split or merged here.
- supportsAn ORCID is claimed by a person, not inferred, so it is the only identity assertion here that a human made.
- weakens4 distinct name form(s) across this row's works. Several unrelated spellings is the usual shape of two people sharing one record.
- weakensAt most 6 distinct institution(s) inside any five-year window. That is more moves than a career usually contains and is the usual shape of a merge.
- supports96% of this row's works sit in its single largest field.
- neutralConfidence is judged on the 8 work(s) this corpus holds, not on the author's whole output. A single-work row carries little evidence either way.