Devansh Arpit

2.2k total citations · 1 hit paper
13 papers, 488 citations indexed

About

Devansh Arpit is a scholar working on Artificial Intelligence, Computer Vision and Pattern Recognition and Statistical and Nonlinear Physics. According to data from OpenAlex, Devansh Arpit has authored 13 papers receiving a total of 488 indexed citations (citations by other indexed papers that have themselves been cited), including 8 papers in Artificial Intelligence, 4 papers in Computer Vision and Pattern Recognition and 3 papers in Statistical and Nonlinear Physics. Recurrent topics in Devansh Arpit's work include Stochastic Gradient Optimization Techniques (6 papers), Adversarial Robustness in Machine Learning (3 papers) and Model Reduction and Neural Networks (3 papers). Devansh Arpit is often cited by papers focused on Stochastic Gradient Optimization Techniques (6 papers), Adversarial Robustness in Machine Learning (3 papers) and Model Reduction and Neural Networks (3 papers). Devansh Arpit collaborates with scholars based in United States, Canada and Germany. Devansh Arpit's co-authors include Aaron Courville, Yoshua Bengio, Stanisław Jastrzȩbski, Maxinder S Kanwal, Nicolas Ballas, Tegan Maharaj, Emmanuel Bengio, Asja Fischer, David Krueger and Simon Lacoste-Julien and has published in prestigious journals such as arXiv (Cornell University), PolyPublie (École Polytechnique de Montréal) and International Conference on Learning Representations.

In The Last Decade

Devansh Arpit

12 papers receiving 474 citations

Hit Papers

align trajectories

What are hit papers?

Hit papers significantly outperform the citation benchmark for their cohort. A paper qualifies if it has ≥500 total citations, achieves ≥1.5× the top-1% citation threshold for papers in the same subfield and year (this is the minimum needed to enter the top 1%, not the average within it), or reaches the top citation threshold in at least one of its specific research topics.

A closer look at memorization in deep networks

2017 344 citations Devansh Arpit, Stanisław Jastrzȩbski et al. Jagiellonian University Repository (Jagiellonian University) profile →

Peers — A (Enhanced Table)

Peers by citation overlap · career bar shows stage (early→late) cites · hero ref

Name	h						Papers	Cites
Devansh Arpit United States	8	336	219	38	32	27	13	488
Pavel Izmailov United States	8	333 1.0×	248 1.1×	13 0.3×	29 0.9×	8 0.3×	11	514
Haojie Lin China	6	219 0.7×	224 1.0×	57 1.5×	23 0.7×	5 0.2×	21	484
Sanjiv Kumar United States	7	267 0.8×	317 1.4×	13 0.3×	34 1.1×	8 0.3×	22	595
Li Guo China	11	254 0.8×	261 1.2×	26 0.7×	25 0.8×	7 0.3×	66	524
Timur Garipov Russia	4	260 0.8×	179 0.8×	13 0.3×	22 0.7×	8 0.3×	9	391
Jin Yu China	11	196 0.6×	295 1.3×	12 0.3×	18 0.6×	25 0.9×	21	512
Shuo Zhou Australia	8	133 0.4×	86 0.4×	14 0.4×	18 0.6×	24 0.9×	20	286
Meizhu Liu United States	11	188 0.6×	180 0.8×	36 0.9×	91 2.8×	6 0.2×	21	488
D. A. Podoprikhin Russia	6	242 0.7×	176 0.8×	11 0.3×	22 0.7×	8 0.3×	11	411
Nabil El Akkad Morocco	17	160 0.5×	411 1.9×	13 0.3×	28 0.9×	11 0.4×	62	613

Countries citing papers authored by Devansh Arpit

Since Specialization

Citations

This map shows the geographic impact of Devansh Arpit's research. It shows the number of citations coming from papers published by authors working in each country. You can also color the map by specialization and compare the number of citations received by Devansh Arpit with the expected number of citations based on a country's size and research output (numbers larger than one mean the country cites Devansh Arpit more than expected).

Fields of papers citing papers by Devansh Arpit

Since Specialization

Physical SciencesHealth SciencesLife SciencesSocial Sciences

This network shows the impact of papers produced by Devansh Arpit. Nodes represent research fields, and links connect fields that are likely to share authors. Colored nodes show fields that tend to cite the papers produced by Devansh Arpit. The network helps show where Devansh Arpit may publish in the future.

Co-authorship network of co-authors of Devansh Arpit

This figure shows the co-authorship network connecting the top 25 collaborators of Devansh Arpit. A scholar is included among the top collaborators of Devansh Arpit based on the total number of citations received by their joint publications. Widths of edges represent the number of papers authors have co-authored together. Node borders signify the number of papers an author published with Devansh Arpit. Devansh Arpit is excluded from the visualization to improve readability, since they are connected to all nodes in the network.

All Works

Sort: Min cites: Since: Top N: Style:

13 of 13 papers shown

1.

Arpit, Devansh, et al.. (2022). Ensemble of Averages: Improving Model Selection and Boosting Performance in Domain Generalization. 8265–8277.

2.

Jastrzȩbski, Stanisław, Stanislav Fort, Devansh Arpit, et al.. (2020). The Break-Even Point on Optimization Trajectories of Deep Neural Networks. arXiv (Cornell University). 3 indexed citations

3.

Jastrzȩbski, Stanisław, Devansh Arpit, Caiming Xiong, et al.. (2020). Catastrophic Fisher Explosion: Early Phase Fisher Matrix Impacts Generalization. arXiv (Cornell University). 4 indexed citations

4.

Rahaman, Nasim, Devansh Arpit, Aristide Baratin, et al.. (2018). On the Spectral Bias of Deep Neural Networks. arXiv (Cornell University). 18 indexed citations

5.

Jastrzȩbski, Stanisław, Zachary Kenton, Devansh Arpit, et al.. (2018). Finding Flatter Minima with SGD. International Conference on Learning Representations. 3 indexed citations

6.

Arpit, Devansh, et al.. (2018). h-detach: Modifying the LSTM Gradient Towards Better Optimization. arXiv (Cornell University). 8 indexed citations

7.

Rahaman, Nasim, Aristide Baratin, Devansh Arpit, et al.. (2018). On the Spectral Bias of Neural Networks. arXiv (Cornell University). 5301–5310. 50 indexed citations

8.

Krueger, David, Nicolas Ballas, Stanisław Jastrzȩbski, et al.. (2017). Deep Nets Don't Learn via Memorization. PolyPublie (École Polytechnique de Montréal). 23 indexed citations

9.

Arpit, Devansh, et al.. (2017). Person Re-identification for Improved Multi-person Multi-camera Tracking by Continuous Entity Association. 566–572. 15 indexed citations

10.

Arpit, Devansh, Stanisław Jastrzȩbski, Nicolas Ballas, et al.. (2017). A closer look at memorization in deep networks. Jagiellonian University Repository (Jagiellonian University). 70. 233–242. 344 indexed citations breakdown →

11.

Zhou, Yingbo, Devansh Arpit, Ifeoma Nwogu, & Venu Govindaraju. (2014). Joint Training of Deep Auto-Encoders. arXiv (Cornell University). 2 indexed citations

12.

Arpit, Devansh, Shuang Wu, Pradeep Natarajan, Rohit Prasad, & Prem Natarajan. (2013). Ridge Regression based classifiers for large scale class imbalanced datasets. 267–274. 7 indexed citations

13.

Arpit, Devansh & Anoop Namboodiri. (2011). Fingerprint feature extraction from gray scale images by ridge tracing. 1–8. 11 indexed citations

Rankless uses publication and citation data sourced from OpenAlex, an open and comprehensive bibliographic database. While OpenAlex provides broad and valuable coverage of the global research landscape, it—like all bibliographic datasets—has inherent limitations. These include incomplete records, variations in author disambiguation, differences in journal indexing, and delays in data updates. As a result, some metrics and network relationships displayed in Rankless may not fully capture the entirety of a scholar's output or impact.

Explore authors with similar magnitude of impact