Product search usually involves a retrieval phase and a ranking phase. Learning-to-rank models improve the ranking phase by learning from customer feedback—and now Amazon scientists have extended that to include retrieval. #ReinforcementLearning

In honor of #IWD2023, we're celebrating the female scientists at Amazon who are making groundbreaking contributions to their fields, breaking stereotypes, and paving the way for future generations of #WomeninSTEM. Meet a few of the notable women:

Explore Amazon researchers’ 40+ accepted papers at this year's @emnlpmeeting, ranging from topics like information extraction, question answering, query rewriting, geolocation, and pun generation. #EMNLP2022 #ConversationalAI #NLProc

"Self-driving cars with #Duckietown", the first vehicle autonomy massive open online course with hands-on learning, is back with a new edition starting November 29th! Hosted on @edXOnline and powered by @NVIDIAEmbedded! Enroll for free! Learn more at: https://t.co/KHOcx6oWX9

Building datasets big enough to train machine learning models is a recurring challenge for Amazon Robotics AI researchers. The solution: a “universal model” able to generalize to virtually any package segmentation task. #ML

A shout-out to Shahriyar Al Mustakim Mitul, he's organising DeepRacer events to promote #machinelearning with students in #Bangladesh! He has created a Facebook Page: https://t.co/L3zidShRWh Please share to help him promote this wonderful initiative! #students #awsdeepracer

#AWSDeepRacer: the most important #AWSreInvent guide out there. What you can see, learn and _win_, all in one place! https://t.co/aNc4rWBtmI #deepracer #MachineLearning #reInvent #aws

Neural MMO Challenge update: imitation learning baseline 2.0 released with a 0.7 top1 ratio. Join the competition without RL experience -> win the Unplugged Prize for offline learning! https://t.co/11hu4BnJzm

Too often, we focus on fixing the robot instead of the environment that it operates in. In this new paper, we do the latter, and show that subtle modifications to the robots’ surroundings make a real difference: https://t.co/qZBEc7bOeW

Look at the speed of these things! They're fully autonomous, cost less than $400 and are faster than anything coming out of universities. Great example of an amateur community solving a hard tech problem and having a blast doing it. Congrats Daniel! (I'm the MC at left)

#ReinforcementLearning #ArtificialIntelligence The printed version of my “Lessons from AlphaZero for Optimal Model Predictive, and Adaptive Control” book has been published; see https://t.co/Hgt8XzvidM An ebook version at

“What happened to Reinforcement Learning research and labs?” Discussion: https://t.co/GLrNtTqQa4

Wohoo! MMO challenge in on! Pass qualifiers for AWS credits. Check out the IJCAI competition for cheat sheet :)

I will be talking about transfer learning and Markov games in #RL with #AWSDeepRacer at Artificial Intelligence for Autonomous Driving - AI4AD Workshop.

Come to the #AWSDeepRacer track at the #Toronto #AWSSummit for some serious #machinelearning powered racing and friendly conversations on how you can get started with machine learning! #AWS

This is the new frontier of Reinforcement learning✨ Build AI agents that learn crucial skills and survive and thrive in large open worlds full of adversaries. ⚔️ Join the Neural MMO Challenge on AIcrowd https://t.co/bq9tPlobpo

I just wrote this blog post about how to train DeepRacer models on EC2 Spot instances. #deepracer #aws #spot #ec2 #machinelearning https://t.co/KK6KWCVj8H

My favorite conference is a small one: The Multi-disciplinary Conference on Reinforcement Learning and Decision Making. It works best if only those with a genuine interest in crossing disciplines attend.

There is something exciting coming up in the AI space in Edmonton: #AIWeek2022, https://t.co/EsXrNU76PM. Events, presentations, workshops, socials. So many people, so much AI!

Hahaha... A real world application of deep #ReinforcementLearning @sunilmallya #AWSDeepRacer