Montreal

Rahul A R

Expert
@rahular

itihasa. A large scale Sanskrit-English translation dataset

83

deepmind-dqn. The original Deepmind Atari 2600 DQN code

32

chess-misc. Misc tools for rStock

28

coref-rl. Rewarding Coreference Resolvers for Being Consistent with World Knowledge

9

varta. Python

8

robust-maml. https://aclanthology.org/2021.findings-acl.106/

7

ellipsis-baselines. A Simple Transfer Learning Baseline for Ellipsis Resolution

4

model-based-coref. Resources for "Model-based annotation of coreference"

4

ramayana. English translation of Valmiki Ramayana by Hari Prasad Shastri

3

PhoneAuth. An Android application built for additional authentication and user verification. The app when installed on the user’s mobile, automatically performs a handshake with a pre-registered computer as and when it comes into Bluetooth range and identifies itself using the IMEI and MAC of the device. Thus, it can uniquely identify the user using his/her mobile device. This is built as a layer on top of the usual username-password protection system (using the Linux PAM APIs). This process happens periodically which means that if the user is not in the vicinity of the computer, it is immediately locked.

3

CriminalTracker. A smart tracker which uses GPS data to predict path trajectories of potential criminals. Won the IBM iCare challenge in 2014.

2

darviz-issues. This repository is meant for raising and monitoring issues related to @ibmdarviz

2

TicTacToe. The classic game as a testbed for AI players

2

rStock. This project modifies Stockfish and plugs in a neural evaluation function and a companion trainer for purely academic purposes. I thank Andrey Kotlarski without whose BPN, this project would be a lot harder to implement.

2

Othello. Java

1

joint-coref-srl. Python

1

2048. Port of Gabriele Cirulli's 2048 to Android (https://github.com/gabrielecirulli/2048)

1

android-eye-tracker. A simple eye tracker for Android

1

rStockTrainer. This is the trainer module for the rStock engine. It trains a new or exisitng NN and saves it future use. We can also test the trained network and find the sqaured error for a perticular training set. Also, the number of training threads and epochs can be set according to needs.

1

neuroph. Neuroph clone with Eclipse compatibility

1

BasicSearch. A simple search engine written in Python to acclimatize myself with various concepts such as k-gram indexing, stemming, etc

1
21
Apply