This video shows the evolution of the parameters (left) and the predictor (right) for a wide 2-layer ReLU neural network (gradient descent on logistic loss). I explain a bit of what's going on in this virtual talk I gave at MSRI (jt. work w/@BachFrancis) https://t.co/GjC8NPNKfX
Learn the concepts of #transferlearning and leverage it in NLP to build incredibly accurate models using popular #fastai library. Also explore #ULMFiT framework - a method to enable transfer learning for any NLP task & achieve great results. https://t.co/1kFjLxjJqd #DeepLearning
Optimizing hyperparameters for machine learning models is a key step in making accurate predictions, as they define characteristics of the model that can impact model accuracy and computational efficiency. #DataScience#MachineLearning https://t.co/JYHs0KiFB4
This paper gives some really nice insights and mathematical depth to what had previously (for us) been “the mystery of squared distance” in revealing the representation of parse trees in deep contextual representations (BERT, ELMo, etc.). Great to read!