{"arxiv_id":"1903.12269","incoming":{"papers":3,"sentences":3,"groups":[{"kind":"background","label":"Cites it as background","papers":2,"sentences":2,"quotes":[{"arxiv_id":"1909.05193","title":"TBT: Targeted Neural Network Attack with Bit Trojan","url":"/paper/1909.05193","section":"Introduction","sentence":"Among all, several major security concerns are adversarial input/example attack [ref] , adversarial parameter attack [ref] and Trojan attack [ref] .","intent_p":0.92,"contests_p":null},{"arxiv_id":"2001.03814","title":"Functional Error Correction for Robust Neural Networks","url":"/paper/2001.03814","section":"Overview of Related Works","sentence":"In [ref] , several security attack methods were tested to find specific error patterns that can cause serious damage to neural networks' performance.","intent_p":0.99,"contests_p":null}]}],"unlabelled_sentences":1},"outgoing":{"papers":11,"sentences":20,"groups":[{"kind":"method_or_tool","label":"Uses the method or tool of","papers":3,"sentences":3,"quotes":[{"arxiv_id":"1308.3432","title":"Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computation","url":"/paper/1308.3432","section":"Quantization and Encoding","sentence":"For training the quantized DNN with non-differential stair-case function (in <ref>), we use the straight-through estimator [ref] as other works [ref] .","intent_p":1.0,"contests_p":null},{"arxiv_id":"1606.06160","title":"DoReFa-Net: Training Low Bitwidth Convolutional Neural Networks with Low Bitwidth Gradients","url":"/paper/1606.06160","section":"Quantization and Encoding","sentence":"For training the quantized DNN with non-differential stair-case function (in <ref>), we use the straight-through estimator [ref] as other works [ref] .","intent_p":1.0,"contests_p":null},{"arxiv_id":"1412.6572","title":"Explaining and Harnessing Adversarial Examples","url":"/paper/1412.6572","section":"Bit-Flip Attack","sentence":"In this work, we perform the BFA utilizing the similar mechanism as FGSM [ref] , which was used to generate adversarial example.","intent_p":0.97,"contests_p":null}]},{"kind":"background","label":"Cites as background","papers":5,"sentences":10,"quotes":[{"arxiv_id":"1412.6572","title":"Explaining and Harnessing Adversarial Examples","url":"/paper/1412.6572","section":"Introduction","sentence":"Adversarial example [ref] (aka., adversarial attack) is a well-known security issue of DNN, which can cause the system malfunction with the magnitude-constrained input noise that mankind cannot discern.","intent_p":0.95,"contests_p":null},{"arxiv_id":"1811.09310","title":"Parametric Noise Injection: Trainable Randomness to Improve Deep Neural Network Robustness against Adversarial Attack","url":"/paper/1811.09310","section":"Introduction","sentence":"Both attack and defense of adversarial example on the input end of DNN has been heavily investigated in the past couple of years [ref] and still be in progress [ref] .","intent_p":0.93,"contests_p":null},{"arxiv_id":"1806.05768","title":"Hardware Trojan Attacks on Neural Networks","url":"/paper/1806.05768","section":"Related Work","sentence":"Neural network parameters have been attacked using different levels of hardware trojans, which require a specific pattern of input to trigger the trojan inside the network [ref] .","intent_p":0.84,"contests_p":null},{"arxiv_id":"1609.03499","title":"WaveNet: A Generative Model for Raw Audio","url":"/paper/1609.03499","section":"Introduction","sentence":"Recently, deep neural networks (DNNs) have demonstrated its great potential of surpassing or close to human-level performance in multiple domains, such as object recognition [ref] , Game AI [ref] , synthetic voice [ref] , neighborhood voting prediction [ref] and etc [ref] .","intent_p":0.98,"contests_p":null},{"arxiv_id":"1508.06576","title":"A Neural Algorithm of Artistic Style","url":"/paper/1508.06576","section":"Introduction","sentence":"Recently, deep neural networks (DNNs) have demonstrated its great potential of surpassing or close to human-level performance in multiple domains, such as object recognition [ref] , Game AI [ref] , synthetic voice [ref] , neighborhood voting prediction [ref] and etc [ref] .","intent_p":0.98,"contests_p":null}]}],"unlabelled_sentences":7},"claims":[],"thresholds":{"intent_p":0.8,"supports_p":0.95,"contests_p":0.7},"method":"Citing sentences come from each paper's LaTeX source; every \\cite marks the sentence and the works it cites. Pith's classifier typed each sentence. A label is shown at intent probability 0.8 or more (0.95 for support), a contest at 0.7 or more. Measured accuracy against two blind model families is at https://pith.science/accuracy. The quoted sentences are the citing authors' own.","held_up":null,"_links":{"html":"https://pith.science/paper/1903.12269"}}