Adam Fisch

Few-Shot Conformal Prediction

Resources for Programming Tutorials

Calibrated Selective Classification

Efficient Conformal Prediction via Cascaded Inference with Expanded Admission

Conformal Prediction Sets with Limited False Positives

COS 429 final project completed by Adam Fisch and Max Shatkhin. Using image processing techniques including Eulerian Video Magnification (see http://people.csail.mit.edu/mrub/vidmag/) and HSV filtering, as well as prediction algorithms such as an HMM, we try to detect anxiety levels of subjects in videos.

Working Hard or Hardly Working: Challenges of Integrating Typology into Neural Dependency Parsers

Intuitive Annotation Tool for Extractive Question Answering (see https://github.com/Varal7/ieturk)

Message Passing Neural Networks for Molecule Property Prediction

A library for transfer learning by reusing parts of TensorFlow models.

Image captioning codebase in pytorch(finetunable cnn in branch "with_finetune";diverse beam search can be found in 'dbs' branch; self-critical training is under my self-critical.pytorch repository.)

Code for Discriminability objective for training descriptive captions(CVPR 2018)

A set of examples around pytorch in Vision, Text, Reinforcement Learning, etc.

PyTorch Tutorial for Deep Learning Researchers

Graph-based and Transition-based dependency parsers based on BiLSTMs

A collection of useful .gitignore templates