1 code implementation • 15 May 2021 • Sunil Thulasidasan, Sushil Thapa, Sayera Dhaubhadel, Gopinath Chennupati, Tanmoy Bhattacharya, Jeff Bilmes
In this work we present a simple, but highly effective approach to deal with out-of-distribution detection that uses the principle of abstention: when encountering a sample from an unseen class, the desired behavior is to abstain from predicting.
Ranked #1 on Out-of-Distribution Detection on CIFAR-100 (using extra training data)
no code implementations • 9 Mar 2021 • Sungwoo Park, Rajan Gupta, Boram Yoon, Santanu Mondal, Tanmoy Bhattacharya, Yong-Chull Jang, Bálint Joó, Frank Winter
Similarly, we find evidence that the $N\pi\pi $ excited state contributes to the correlation functions with the vector current, consistent with the vector meson dominance model.
High Energy Physics - Lattice High Energy Physics - Phenomenology
no code implementations • 18 Jan 2021 • Tanmoy Bhattacharya, Vincenzo Cirigliano, Rajan Gupta, Emanuele Mereghetti, Boram Yoon
Using the excited state spectrum from fits to the two-point function, we find $d_n^\Theta$ is small, $|d_n^\Theta| \lesssim 0. 01 \overline \Theta e$ fm, whereas for the proton we get $|d_p^\Theta| \sim 0. 02 \overline \Theta e$ fm.
High Energy Physics - Lattice High Energy Physics - Phenomenology
no code implementations • 1 Jan 2021 • Sunil Thulasidasan, Sushil Thapa, Sayera Dhaubhadel, Gopinath Chennupati, Tanmoy Bhattacharya, Jeff Bilmes
In this work we present a simple, but highly effective approach to deal with out-of-distribution detection that uses the principle of abstention: when encountering a sample from an unseen class, the desired behavior is to abstain from predicting.
no code implementations • 23 Dec 2020 • Cristina Garcia-Cardona, M. Giselle Fernández-Godino, Daniel O'Malley, Tanmoy Bhattacharya
Simulation of the crack network evolution on high strain rate impact experiments performed in brittle materials is very compute-intensive.
no code implementations • 3 Dec 2020 • Tanmoy Bhattacharya, Alexander J. Buser, Shailesh Chandrasekharan, Rajan Gupta, Hersh Singh
We provide strong evidence that the asymptotically free (1+1)-dimensional non-linear O(3) sigma model can be regularized using a quantum lattice Hamiltonian, referred to as the "Heisenberg-comb", that acts on a Hilbert space with only two qubits per spatial lattice site.
High Energy Physics - Lattice Strongly Correlated Electrons High Energy Physics - Theory Nuclear Theory Quantum Physics
no code implementations • 24 Nov 2020 • Santanu Mondal, Rajan Gupta, Sungwoo Park, Boram Yoon, Tanmoy Bhattacharya, Bálint Joó, Frank Winter
Our final results, in the $\overline{\rm MS}$ scheme at 2 GeV, are $\langle x \rangle_{u-d} = 0. 160(16)(20)$, $\langle x \rangle_{\Delta u-\Delta d} = 0. 192(13)(20)$ and $\langle x \rangle_{\delta u-\delta d} = 0. 215(17)(20)$, where the first error is the overall analysis uncertainty assuming excited-state contributions have been removed, and the second is an additional systematic uncertainty due to possible residual excited-state contributions.
High Energy Physics - Lattice
no code implementations • 10 Sep 2020 • Sayera Dhaubhadel, Jamaludin Mohd-Yusof, Kumkum Ganguly, Gopinath Chennupati, Sunil Thulasidasan, Nicolas W. Hengartner, Brent J. Mumphrey, Eric B. Durbin, Jennifer A. Doherty, Mireille Lemieux, Noah Schaefferkoetter, Georgia Tourassi, Linda Coyle, Lynne Penberthy, Benjamin H. McMahon, Tanmoy Bhattacharya
We demonstrate an abstaining classifier in a multitask setting for classifying cancer pathology reports from the NCI SEER cancer registries on six tasks of interest.
no code implementations • 28 Jun 2020 • Alexander J. Buser, Tanmoy Bhattacharya, Lukasz Cincio, Rajan Gupta
Recently, Singh and Chandrasekharan showed that fixed points of the non-linear O(3) sigma model can be reproduced near a quantum phase transition of a spin model with just two qubits per lattice site.
Quantum Physics High Energy Physics - Lattice
2 code implementations • 27 May 2019 • Sunil Thulasidasan, Tanmoy Bhattacharya, Jeff Bilmes, Gopinath Chennupati, Jamal Mohd-Yusof
In the case of unstructured (arbitrary) label noise, abstention during training enables the DAC to be used as an effective data cleaner by identifying samples that are likely to have label noise.
2 code implementations • NeurIPS 2019 • Sunil Thulasidasan, Gopinath Chennupati, Jeff Bilmes, Tanmoy Bhattacharya, Sarah Michalak
In this work, we discuss a hitherto untouched aspect of mixup training -- the calibration and predictive uncertainty of models trained with mixup.
Ranked #1 on Out-of-Distribution Detection on STL-10
no code implementations • ICLR 2019 • Sunil Thulasidasan, Tanmoy Bhattacharya, Jeffrey Bilmes, Gopinath Chennupati, Jamal Mohd-Yusof
We introduce the deep abstaining classifier -- a deep neural network trained with a novel loss function that provides an abstention option during training.
no code implementations • 29 Apr 2015 • Hyejin Youn, Logan Sutton, Eric Smith, Cristopher Moore, Jon F. Wilkins, Ian Maddieson, William Croft, Tanmoy Bhattacharya
How universal is human conceptual structure?