Published incodeburstA hypothetical AI for segregating responses based on individual stimuliIn the last two weeks, I came across a couple of interesting papers on vision-understanding via Deep Learning. Lets briefly look at both:Apr 18, 2018Apr 18, 2018
Published incodeburstUnderstanding Few-shot intelligence as a Meta-Learning ProblemMeta learning is learning to learn. Usually applied for hyperparameter tuning, recent applications have started focussing on few-shot…Feb 12, 2018Feb 12, 2018
Published incodeburstUnderstanding AttnGAN: Text-to-Image convertorText-to-image conversion has always fascinated me, and the recent AttnGAN paper caught my attention. In this post I try to provide an…Jan 27, 20182Jan 27, 20182
Overfitting and Human BehaviorAlgorithms to Live By draws interesting parallels between familiar algorithms and human psychology. One of the more interesting chapters…Jan 24, 2018Jan 24, 2018
Published incodeburstUnderstanding R-Net: Microsoft’s ‘superhuman’ reading AIThey said that AIs developed by Microsoft and Alibaba beat humans at reading. While not entirely accurate, models developed by these…Jan 19, 2018Jan 19, 2018
Spirituality for the practicalSpirituality means different things to different people. Even then, one common theme across all interpretations is an association to God…Jan 15, 2018Jan 15, 2018
Published incodeburstAdversarial patches for CNNs explainedWhile many articles in the past few weeks have talked about the recent Adversarial Patchs, no one has ventured into explaining how they…Jan 9, 2018Jan 9, 2018
First time with Kaggle: A ConvNet to classify toxic comments with KerasWork has been slow in the first week of the year, so I decided to try my hand at a Kaggle competition for the first time (yeah I know I am…Jan 9, 20181Jan 9, 20181
