Published in Towards AI·PinnedGradient Descent OptimizationGradient Descent Optimization Algorithms — Most deep learning algorithms train by optimizing some kind of objective (or loss) function. We typically try to either maximize or minimize these functions. Gradient-based optimization is the most extensively used technique for training neural networks. In this article, we will see the well-known Gradient Descent algorithm. We will discuss…Machine Learning14 min readMachine Learning14 min read
PinnedUnderstanding Markov Decision Process (MDP)Towards Training Better Reinforcement Learning Agents — In this article, we’ll be discussing the objective using which most of the Reinforcement Learning (RL) problems can be addressed— a Markov Decision Process (MDP) is a mathematical framework used for modeling decision-making problems where the outcomes are partly random and partly controllable. We’ll discuss MDPs in greater detail as…Artificial Intelligence10 min readArtificial Intelligence10 min read
Published in Towards Data Science·PinnedIntriguing Properties of Neural NetworksHow do Neural Nets Work? — Neural Networks have been eminent among machine learning algorithms for more than a decade. They have achieved wonders in numerous tasks in various domains. However, there have always been questions about how they actually work. Neural Networks can fit any training data. It tries to learn a mapping between the…Deep Learning13 min readDeep Learning13 min read
Published in Towards Data Science·PinnedMember-onlyQRNN: A Potential Competitor to the TransformerTraining Faster RNNs with Quasi-RNN — Recurrent Neural Networks (RNNs) have been in the sequence modeling business for a long time. But RNNs are slow; they process one token at a time. Moreover, the recurrent architecture adds a limitation of fixed-length encoding vectors for the complete sequence. …Artificial Intelligence7 min readArtificial Intelligence7 min read
Published in Towards Data Science·PinnedMember-onlyA Comprehensive Guide to Generative Adversarial Networks (GANs)Generating Meaningful Data from Noise — This (GANs), and the variations that are now being proposed is the most interesting idea in the last 10 years in ML, in my opinion — Yann LeCun (VP & Chief AI Scientist at Facebook) in his Quora Session GANs are one of the latest wonders of Deep Learning. These…Deep Learning7 min readDeep Learning7 min read
Published in The Startup·Jan 26, 2021Member-onlyMy 2-Time Google Foobar ExperienceHow I did at the Google Foobar Challenge, Twice! — Foobar is “said to be” Google’s secret hiring challenge, and it’s really exciting. It is an invite-only challenge, so “you have to be chosen.” The picture above is how I received my invite. In this article, I’ll be sharing my experience of the foobar challenge. …Programming6 min readProgramming6 min read
Published in Towards Data Science·Jan 12, 2021Member-onlyGPT-3 ExplainedUnderstanding Transformer-Based Self-Supervised Architectures — In this article, we’ll be discussing the renowned GPT-3 model proposed in the paper “Language Models are Few-Shot Learners” by OpenAI. It is the successor of GPT-2, which has a very similar architecture to that of GPT-3. If you’re unaware of GPT-2, consider giving my article on GPT-2 a read…Artificial Intelligence11 min readArtificial Intelligence11 min read
Published in Towards Data Science·Dec 30, 2020Member-onlyDynamic Programming in RLTowards Training Better Reinforcement Learning Agents — Dynamic Programming is a mathematical optimization approach typically used to improvise recursive algorithms. It basically involves simplifying a large problem into smaller sub-problems. There are two properties that a problem must exhibit to be solved using dynamic programming: Overlapping Subproblems Optimal Substructure We’ll be discussing ‘Planning in RL’ using dynamic…Artificial Intelligence8 min readArtificial Intelligence8 min read
Published in Towards Data Science·Dec 1, 2020Member-onlyLongformer: The Long-Document TransformerUnderstanding Transformer-Based Self-Supervised Architectures — Transformer-based language models have been leading the NLP benchmarks lately. Models like BERT, RoBERTa have been state-of-the-art for a while. However, one major drawback of these models is that they cannot “attend” to longer sequences. For example, BERT is limited to a max of 512 tokens at a time. To…Artificial Intelligence7 min readArtificial Intelligence7 min read
Published in Towards Data Science·Nov 13, 2020Member-onlyOptimizing Model Training with TensorFlow ProfilerOptimizing GPU Performance with TensorFlow — We want our models to train real fast. We use GPUs to make operations execute faster. However, it is possible that even after speeding up the computations, the model may have inefficiencies in the pipeline itself, and thus, may train slower. …Artificial Intelligence9 min readArtificial Intelligence9 min read
