Scribd
Upload
142 views1 page

Arslan Tariq S CV PDF

Ritika Agarwal is pursuing an integrated B.Tech + M.Tech in Information Technology from ABV Indian Institute of Information Technology and Management. She has work experience as a research intern developing an adaptive ANN algorithm using disturbance observer and EKF, and as a machine learning intern developing AI models for computer vision. Some of her projects include a lipreading model using CNN, GRU, attention and CTC loss, and models for text summarization, landmark identification, and toxic comment classification. Her skills include Python, C++, machine learning algorithms and deep learning frameworks like TensorFlow and Keras.

Uploaded by

Ritika Agarwal
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
142 views1 page

Arslan Tariq S CV PDF

Ritika Agarwal is pursuing an integrated B.Tech + M.Tech in Information Technology from ABV Indian Institute of Information Technology and Management. She has work experience as a research intern developing an adaptive ANN algorithm using disturbance observer and EKF, and as a machine learning intern developing AI models for computer vision. Some of her projects include a lipreading model using CNN, GRU, attention and CTC loss, and models for text summarization, landmark identification, and toxic comment classification. Her skills include Python, C++, machine learning algorithms and deep learning frameworks like TensorFlow and Keras.

Uploaded by

Ritika Agarwal
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 1

Ritika Agarwal

ritikaagarwal646@gmail.com | +919899308916

EDUCATION EXPERIENCE
ABV INDIAN INSTITUTE NON-LINEAR DYNAMICS IN THE PRESENCE OF DISTURBANCE
OF INFORMATION WITH ANN | RESEARCH INTERN
TECHNOLOGY AND July 2019 – Present | Remote Work|NSUT,New Delhi
MANAGEMENT • Working remotely as a research intern to develop an algorithm for adaptation
INTEGRATED B.TECH + M.TECH IN of ANNs in the presence of disturbance and noise.
INFORMATION TECHNOLOGY • The algorithm uses disturbance observer and EKF.
Jun 2016 - May 2021 | Gwalior, India
CGPA: 7.48/10 (Upto 7th Semester) INDPULSE TECHNOLOGIES | MACHINE LEARNING INTERN
June 2019 – July 2019 | Remote Work
MODERN INTERNATIONAL • Worked as a machine learning intern and developed AI models for computer
SCHOOL vision use cases.
XII (CBSE)
April 2016 | New Delhi, India
Percentage: 92.8% PROJECTS
X (CBSE)
April 2013 | New Delhi, India BTECH THESIS PROJECT May 2019 – Present | LRNeuNet
CGPA:9.4 Devised an attention based deep architecture for lipreading from multitudinous sized
videos.The model uses spatio temporal CNN ,Bi-GRU and an attention layer along
with CTC loss to predict the spoken sentences by mapping the silent lip movements.
SKILLS The proposed architecture achieved a word error rate and character error rate of
2.7% and 1.2% respectively on the overlapped speakers scenario on the GRID corpus.
PROGRAMMING
•Python • C++ • Data Structures
• Design and Analysis of Algorithms TEXT SUMMARISATION June 2019
Made a seq-seq model through tensorflow for text summarization.The summarization
MACHINE LEARNING model uses a bidirectional RNN, Bahdanau Attention and Beam Search and is applied
• Convolutional Neural Network on 3 different datasets taken from kaggle.
• Text Summarization
• Sequence Labelling LANDMARK IDENTIFICATION June 2019
•Text Classification Devised a model which can predict landmark labels,for example it can distinguish
• Image Recognition between a common waterfall and the Niagra waterfalls,and identify it as such, directly
• Computer Vision from image pixels.The model uses ResNet50 along with Global maxpooling , with
• Natural Language Processing pretrained weights on ImageNet and Adam optimizer.

PYTHON PACKAGES PREVENT OVERFITTING May2019


Developed a solution to kaggle’s Don’t OverfitII competition. The goal was to prevent
•Tensorflow• NumPy •Pandas •NLTK
overfitting. The model selected relevant features through scikits RFECV with L1
•Scikit-Learn • Keras •SpaCy
regularization. The input dataset was split into train and validation dataset using
• OpenCV
Stratified kfold. A 1 hidden layer neural network was used through keras.
The model achieved an accuracy of 87.2%.
LINKS TOXIC COMMENT CLASSIFICATION Jan 2019
LinkedIn:// ritika-agarwal-b2856912b Devised a model for categorizing comments within various classes of toxicity ,using
GitHub:// RitikaAg TFIDF vectorizer and logistic regression for the Toxic Comment Classification
HackerRank:// @ritika0 7 Challenge on kaggle.
The model achieved an accuracy of 97%.

COURSEWORK MOOCS
UNDERGRADUATE •Improving Deep Neural Networks: Hyperparameter tuning, Regularization and
•Artificial Intelligence Optimization|July 2018 •Structuring Machine Learning Projects|July 2018 •Neural
•Object Oriented Programming Networks and Deep Learning|June 2018
•Data Structures
•Design and Analysis of Algorithms
•Database Management Systems

You might also like