1 code implementation • NeurIPS 2020 • Jaehyung Kim, Youngbum Hur, Sejun Park, Eunho Yang, Sung Ju Hwang, Jinwoo Shin
While semi-supervised learning (SSL) has proven to be a promising way for leveraging unlabeled data when labeled data is scarce, the existing SSL algorithms typically assume that training class distributions are balanced.
1 code implementation • ICLR 2021 • Jaeho Lee, Sejun Park, Sangwoo Mo, Sungsoo Ahn, Jinwoo Shin
Recent discoveries on neural network pruning reveal that, with a carefully chosen layerwise sparsity, a simple magnitude-based pruning achieves state-of-the-art tradeoff between sparsity and performance.
1 code implementation • ICLR 2020 • Sejun Park, Jaeho Lee, Sangwoo Mo, Jinwoo Shin
Magnitude-based pruning is one of the simplest methods for pruning neural networks.
1 code implementation • NeurIPS 2021 • Jongheon Jeong, Sejun Park, Minkyu Kim, Heung-Chang Lee, DoGuk Kim, Jinwoo Shin
Randomized smoothing is currently a state-of-the-art method to construct a certifiably robust classifier from neural networks against $\ell_2$-adversarial perturbations.
1 code implementation • 6 Mar 2023 • Hankook Lee, Jongheon Jeong, Sejun Park, Jinwoo Shin
To enable the joint training of EBM and CRL, we also design a new class of latent-variable EBMs for learning the joint density of data and the contrastive latent variable.
1 code implementation • NeurIPS 2020 • Jaeho Lee, Sejun Park, Jinwoo Shin
The second result, based on a novel variance-based characterization of OCE, gives an expected loss guarantee with a suppressed dependence on the smoothness of the selected OCE.
no code implementations • 16 Dec 2014 • Sejun Park, Jinwoo Shin
The max-product {belief propagation} (BP) is a popular message-passing heuristic for approximating a maximum-a-posteriori (MAP) assignment in a joint distribution represented by a graphical model (GM).
no code implementations • 6 Apr 2017 • Sejun Park, Yunhun Jang, Andreas Galanis, Jinwoo Shin, Daniel Stefankovic, Eric Vigoda
The Gibbs sampler is a particularly popular Markov chain used for learning and inference problems in Graphical Models (GMs).
no code implementations • 12 Mar 2017 • Sejun Park, Eunho Yang, Jinwoo Shin
Learning parameters of latent graphical models (GM) is inherently much harder than that of no-latent ones since the latent variables make the corresponding log-likelihood non-concave.
no code implementations • NeurIPS 2015 • Sungsoo Ahn, Sejun Park, Michael Chertkov, Jinwoo Shin
Max-product Belief Propagation (BP) is a popular message-passing algorithm for computing a Maximum-A-Posteriori (MAP) assignment over a distribution represented by a Graphical Model (GM).
no code implementations • 14 May 2019 • Sejun Park, Eunho Yang, Se-Young Yun, Jinwoo Shin
Our contribution is two-fold: (a) we first propose a fully polynomial-time approximation scheme (FPTAS) for approximating the partition function of GM associating with a low-rank coupling matrix; (b) for general high-rank GMs, we design a spectral mean-field scheme utilizing (a) as a subroutine, where it approximates a high-rank GM into a product of rank-1 GMs for an efficient approximation of the partition function.
no code implementations • ICLR 2021 • Sejun Park, Chulhee Yun, Jaeho Lee, Jinwoo Shin
In this work, we provide the first definitive result in this direction for networks using the ReLU activation functions: The minimum width required for the universal approximation of the $L^p$ functions is exactly $\max\{d_x+1, d_y\}$.
no code implementations • 26 Oct 2020 • Sejun Park, Jaeho Lee, Chulhee Yun, Jinwoo Shin
It is known that $O(N)$ parameters are sufficient for neural networks to memorize arbitrary $N$ input-label pairs.
no code implementations • ICML Workshop AML 2021 • Jongheon Jeong, Sejun Park, Minkyu Kim, Heung-Chang Lee, DoGuk Kim, Jinwoo Shin
Randomized smoothing is currently a state-of-the-art method to construct a certifiably robust classifier from neural networks against $\ell_2$-adversarial perturbations.
no code implementations • 19 Sep 2022 • Sejun Park, Umut Şimşekli, Murat A. Erdogdu
In this paper, we propose a new covering technique localized for the trajectories of SGD.
no code implementations • 29 Sep 2022 • Alireza Mousavi-Hosseini, Sejun Park, Manuela Girotti, Ioannis Mitliagkas, Murat A. Erdogdu
We further demonstrate that, SGD-trained ReLU NNs can learn a single-index target of the form $y=f(\langle\boldsymbol{u},\boldsymbol{x}\rangle) + \epsilon$ by recovering the principal direction, with a sample complexity linear in $d$ (up to log factors), where $f$ is a monotonic function with at most polynomial growth, and $\epsilon$ is the noise.
no code implementations • 31 Jan 2023 • Wonyeol Lee, Sejun Park, Alex Aiken
For a neural network with bias parameters, we first prove that the incorrect set is always empty.
no code implementations • 19 Sep 2023 • Namjun Kim, Chanho Min, Sejun Park
We next prove a lower bound on $w_{\min}$ for uniform approximation using general activation functions including ReLU: $w_{\min}\ge d_y+1$ if $d_x<d_y\le2d_x$.
no code implementations • 26 Jan 2024 • Yeachan Park, Geonho Hwang, Wonyeol Lee, Sejun Park
In this work, we analyze the expressive power of neural networks under a more realistic setup: when we use floating-point numbers and operations.