PinnedNiranjan RajeshinBits and NeuronsWhy Adversarial Attacks are important in the path to human-like-AIAn outline of how adversarial attacks arise, why they are problematic and how they reveal important insights about the state of AI today.Jul 8Jul 8
PinnedNiranjan RajeshinBits and NeuronsCNNs vs Vision Transformers — Biological Computer Vision (3/3)The third article in Biological Computer Vision. We discuss the differences of the two state of the art architectures in computer vision.Jan 12, 20231Jan 12, 20231
Niranjan RajeshinBits and NeuronsNeural Manifolds — Linear Algebra and Topology in NeuroscienceA look at how vector and topological spaces play a role in deciphering complex neural data.Aug 26, 2023Aug 26, 2023
Niranjan RajeshinBits and NeuronsModern Brain-Inspired CNNsAn exploration of recent attempts by MIT’s DiCarlo Lab to make CNNs more ‘Brain-Like’ and its consequences.Aug 3, 20231Aug 3, 20231
Niranjan RajeshinBits and NeuronsVision Transformers from a Biological Standpoint — Biological Computer Vision (2/3)In this article, I explain how Vision Transformers work while comparing it with the biological mechanisms they take inspiration from.Sep 14, 20221Sep 14, 20221
Niranjan RajeshinBits and NeuronsCNNs from a Biological Standpoint — Biological Computer Vision (1/3)Convolutional Neural Networks are the industry standard when it comes to Computer Vision. How similar is it to the human visual system?Aug 15, 2022Aug 15, 2022
Niranjan RajeshinBits and NeuronsDo infants hold the key to developing superior Artificial Intelligence?On why it is important to pay attention to developmental scientists to develop better unsupervised learning algorithms.Aug 7, 2022Aug 7, 2022
Niranjan RajeshinBits and NeuronsMemory in Human and Artificial ComputationHuman and artificial computations rely on memoization to speed up their algorithms. How exactly does this happen?Jun 6, 2022Jun 6, 2022
Niranjan RajeshinBits and NeuronsVideo Games might be the key to understanding Human Learning and getting closer to Human-like AIA review of recent work that tries that starts to solve the problem of bringing Artificial Learning closer to Human Learning.May 30, 20224May 30, 20224