Archit Sakhadeo Email & Phone Number

Toronto, Ontario, Canada

Edmonton, Alberta, Canada

I was advised by Prof. Adam White and Prof. Alona Fyshe. My focus was on representation learning, reinforcement learning (RL), automated hyperparameter tuning, and deploying RL in the real world.Along with Prof. Adam White, Prof. Martha White, and some MSc and PhD students, I worked on tackling a real-world industrial control task. The task was to optimize a water treatment plant with reinforcement learning to produce safer, cleaner drinking water, lower energy emissions, and reduce chemical usage.Helped design and conduct experiments by implementing several RL agents and simulated environments that reflect the features of the water treatment process. Helped design a methodology for offline hyperparameter tuning for the deployment of online RL agents through the use of logs of historical data from real-world applications. Demonstrated the efficacy of this approach through empirical studies.

Edmonton, Alberta, Canada

I was a teaching assistant for the following courses:1) Empirical Reinforcement Learning (graduate course) - CMPUT 607 Taught by: Prof. Adam White2) Machine Learning (undergraduate course) - CMPUT 296 Taught by: Prof. Martha White3) Foundations of Computation (undergraduate course) - CMPUT 174Taught by: Prof. Joerg Sander, Prof. Sadaf Ahmed, and Prof. Geoff Hollis----------------------------------------------------------------------------------------My love for teaching emerged through being a TA for the Machine Learning course.During all these courses, I took on the following roles:1) I conducted tutorials on convex optimization.2) I guided students to plan their research projects, helped them design experiments, and gave regular feedback on their progress.3) I conducted labs and office hours to solve student questions about the course material. I simplified and taught complex concepts during the labs.4) I designed programming assignments.5) I graded the quizzes and exams.

Contributed to the efforts in building a complete framework for constructing a commonsense knowledge base from several sources on the internet. Constructed a high-quality knowledge base, of 2.26 million salient properties for 80000 concepts, which is publicly available as a research resource. Improved recall by 83% and quality (measured through a human survey) over state-of-the-art knowledge bases.

By conducting MEG experiments on participants reading a stimuli text corpus, the existence of a soft correlation between the P600 event-related potential and the absolute difference of semantic associations was proven in order to study how the brain disambiguates prepositional phrase attachments. Corpus-derived co-occurrence frequencies of stimuli words were used as a measure of semantic associations.

Developed an extractive text summarizer. Designed a hybrid approach of using keyword frequencies (statistical approach) with automatically generated entity relationships (semantic approach) to combine the strengths and ameliorate the weaknesses of both approaches to generate better summaries. Achieved 16.44% increase in recall and 2.17% increase in F-score than the next best technique. A survey on 94 participants also suggested that our method's summaries were more human-like than the tested methods' summaries.

Along with a team, designed a data processing pipeline to run on a high performance compute cluster to synthesize images from radio interferometric data from the Giant Metrewave Radio Telescope (GMRT). It helped to reduce the synthesis time to under a week from 2+ years. It has enabled newer cosmological insights and the creation of one of the world's largest catalogues for radio astronomy images. Our work was presented at the 30th General Assembly of The International Astronomical Union in Vienna, the 36th Annual Meeting of The Astronomical Society of India, and is accepted at ADASS 2018.

TEDxPICT was founded with the intention of promoting social good, thoughtfulness, and discussion. Started as an informal college-level discussion group, it is now an annual city-level event that hosts talks by people who transformed extraordinary ideas into reality. A global platform is provided to the local community as these talks go online. As a curator, my job entailed ideating on the theme and working alongside credited speakers in developing the content of their talk.

Master Of Science - Msc (Thesis), Computing Science

Thesis: No more pesky hyperparameters: Offline hyperparameter tuning for reinforcement learning Advisors: Prof. Adam White, Prof. Alona.

Bachelor’S Degree, Computer Engineering

Education record