AND D

By M. Wiggins

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Natural Language Processing

Citations in the corpus

Semantic Image Synthesis With Spatially-Adaptive Normalization (2019, neutral): “Earlier methods compute the output image by stitching pieces from a database of images [3, 14].”
A literature survey of low‐rank tensor approximation techniques (2013, neutral): “There is also a Python implementation of the TT-toolbox called ttpy [204].”
Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization (2016, neutral): “Human attention maps are compared to Grad-CAM visualizations for the VQA model from [38] on 1374 val question-image (QI) pairs from [3] using the rank correlation evaluation protocol as in [9].”
Recursive partitioning for heterogeneous causal effects (2015, neutral): “[25] consider a model with the outcome linear in the covariates and the interaction with the treatment variable.”
Class-Incremental Continual Learning Into the eXtended DER-Verse (2022, neutral): “• Learning a Unified Classifier Incrementally via Rebalancing (LUCIR) [31] is a rehearsal strategy proposing several modifications that promote separation in feature space and result in a more harmonized incrementally-learned classifier; • Greedy Sampler and Dumb learner (GDumb) [13] is an experimen”
A contextual-bandit approach to personalized news article recommendation (2010, neutral): “Furthermore, a significant nu mber of visitors are likely to be entirely new with no historic al consumption record whatsoever; this is known as a cold-start situation [21].”
PowerPlay: Training an Increasingly General Problem Solver by Continually Searching for the Simplest Still Unsolvable Problem (2011, neutral): “While kids keep inventing new problems for themselves, they move t hrough remarkable developmental stages [23, 2, 11].”
Reward Augmented Maximum Likelihood for Neural Structured Prediction (2016, positive): “A canonical example is large margin structured prediction [38, 40, 13], which explicitl y uses supervision and considers an upper bound surrogate for task loss.”
Deep learning for detecting robotic grasps (2013, neutral): “While some works [10, 46] simply attempt to estimate the poses of known objects and then apply full-model grasping algorithms based on these results, others avoid this assumption, functioning on novel objects which the algorithm has not seen before.”
Testing for Homogeneity with Kernel Fisher Discriminant Analysis (2007, neutral): “2 Speaker verification We conducted experiments in a speaker verification task [3], on a subset of 8 female speakers using data from the NIST 2004 Speaker Recognition Evaluation.”
Learning with Hierarchical-Deep Models (2013, neutral): “We also place Gamma priors over concentration parameters as in [38].”
nuScenes: A Multimodal Dataset for Autonomous Driving (2019, neutral): “5 15k 15k 0 15k 200k No No No Karlsruhe AS lidar [40] 2018 - 2 0 20k 0 20k 475k - - KAIST [11] 2018 - - 8.”
DAPPLE: a pipelined data parallel approach for training large models (2020, neutral): “, NLP [2], Internet scale E-commerce search/recommendation systems [3], [4]) have billions of parameters, demanding tens to hundreds of GBs of device memory for training.”
Compressing Visual-linguistic Model via Knowledge Distillation (2021, neutral): “, syntactic and co-reference relation of input tokens [10, 31].”
Flow-based generative models for Markov chain Monte Carlo in lattice field theory (2019, neutral): “Complex coupling layers have been implemented [39, 52], as have generalized convolutions [53, 54] and transformations with continuous dynamics that are not dependent on restricted coupling layers [55].”
A Unified Convergence Analysis of Block Successive Minimization Methods for Nonsmooth Optimization (2012, neutral): “In fact, as we will see later in this paper, successively optimizing an approximat e version of the original objective is the key idea of many important algorithms such as the concave-conve x procedure [33], the EM algorithm [10], the proximal minimization algorithm [2], to name a few.”
Bypassing the Kohn-Sham equations with machine learning (2016, positive): “D DFT calculations for ML models are performed with the Quantum ESPRESSO code[36] using the PBE exchange-correlation functional[33] and pro-jector augmented waves (PAWs)[34, 35] with Troullier-Martin pseudization for describing the ionic cores[45].”
B-physics anomalies: a guide to combined explanations (2017, unknown): “In particular, we take into account the LFU tests in the charged-current semi-leptonic observables Rτ` D(∗) and R b→c, global fits of b→ sμμ processes (including the LFU ratios R μe K(∗) and the angular observables) along the direction ∆C 9 = −∆C μ 10 [36–42], and limits on B(B → K∗νν̄) [43].”