Ali Chaudhry
Ali is a self-taught programmer, currently doing a PhD in Artificial Intelligence and Education at University College London. His research focuses on applying human-centered learning science theories on Deep Reinforcement Learning algorithms. These days he's obsessed with GANs and Variational Autoencoders.
Ali is also running a charity named Renaissance which operates number of charity schools for the disadvantaged communities in Pakistan. His schools focus on quality education where technology is thoroughly integrated in classrooms and day-to-day operations. He has also launched a number of AI-based technology products in Pakistan, disrupting the Agriculture sector through Qurbani App and Education sector through Naseem Education.
Talks I've Given
-
Task-Oriented Query Reformulation with Reinforcement Learning
Featuring Ali Chaudhry
This meetup will be a Journal Club with a discussion on the following paper: Task-Oriented Query Reformulation with Reinforcement Learning.
Search engines play an important role in our everyday lives by assisting us in finding the information we need. When we input a complex query, however,...
discovery exploration reinforcement-learning -
Synthesizing Programs for Images using Reinforced Adversarial Learning
Featuring Ali Chaudhry
The code for Reinforcement Learning agents trained in this paper was open sourced last month here:
exploration data reinforcement-learning -
Reinforcement Learning & Creative Applications
Featuring Ali Chaudhry
Reinforcement Learning (RL) has been a key part of the huge growth and advances in AI over recent years.
Combined with deep neural networks, RL has enabled Google Deepmind's DQG to play Atari games, and was a key part of the AlphaGo system that mastered the notoriously difficult 3000-year old...
data atari reinforcement-learning art -
Meta Reinforcement Learning: The Power of Generalizing RL Algorithms
Featuring Ali Chaudhry
Meta Reinforcement Learning is like Transfer Learning in Reinforcement Learning algorithms. Meta Reinforcement Learning has recently gained a lot of popularity by producing impressive results on a number of different tasks. It was extensively discussed at ICLR'19 this year.
In this talk, Ali...
discovery data meta-reinforcement-learning reinforcement-learning -
Meta Reinforcement Learning: The Power of Generalizing RL Algorithms
Featuring Ali Chaudhry
In this talk, Ali Chaudhry will share the power of Meta Reinforcement Learning algorithms - One of the most active areas of research in AI Labs around the world.
meta-reinforcement-learning algorithms data ethics ai -
Lightning Talk: Quality Education for 25 Million Out of School Kids
Featuring Ali Chaudhry
Pakistan has 6th largest population of the world with more than 25 million out of school children. More than 50 million children study in government schools where the quality of education is below average with no focus on developing 21st Century Skills.
diversity tech-for-good closing-skills-gap -
Round Table Sessions: Responsible AI
Featuring Ali Chaudhry
Join Ali in Esc for a Round Table Session on responsible AI.
diversity roundtable ethics data ai -
Monte Carlo Tree Search in Reinforcement Learning
Featuring Ali Chaudhry
Monte Carlo Tree Search has proven to be one of the most productive techniques in Reinforcement Learning. It has produced state of the results in problems with humongous state spaces like Chess and Go. Ali will present the theory behind this Monte Carlo Tress Search and share its use cases.
data reinforcement-learning -
Reinforcement Learning Basics: Modelling your Algorithms
Featuring Ali Chaudhry
In this talk, Ali Chaudhry will share the choices that one has to make in modelling a Reinforcement Learning algorithm.
discovery data artificial-intelligence reinforcement-learning