This is our first time participating in any Alexa competition and we are excited to tackle the problems facing Alexa in the domains of cooking and do-it-yourself tasks.
Team Maruna is a group of graduate students from the University of Massachusetts at Amherst. This is our first time participating in any Alexa competition and we are excited to tackle the problems facing Alexa in the domains of cooking and do-it-yourself tasks.
We are advised by Prof. Hamed Zamani and collaborate with Prof. Mohit Iyyer and Prof. W. Bruce Croft, who work on various aspects of information retrieval and natural language processing. We aim to leverage existing state-of-the-arts in conversation AI, enhance them further by identifying gaps, push boundaries of multi-modal conversational information seeking systems, and improve overall customer experience while using our taskbot. We focus on both engineering and research problems because everything impacts our customers. Our vision is to create a taskbot that is reliable, friendly, easy to navigate, and gets the job done.
Amit Hattimare - Team leader
I am pursuing a Master's in Computer Science with Data Science concentration. Prior to this I have 4 years of software development experience at Amazon Retail where I was part of the Fulfillment Options team that handles inventory movement in and out of warehouses. I have experience building and maintaining scalable solutions using AWS. My research interests are in conversation retrieval systems, question answering, text generation, multi-domain dialogue state tracking, and multi-language NLP. I have worked on paper replication for SOTA BigBird [2020, Zaheer et al] where I conducted an independent literature review and verified the proposed model's performance on some NLP tasks. I have designed an intent classification model specific to task-oriented dialogue systems (TODS) inspired by this challenge and currently working on handling QA in TODS. I also enjoy integrating research components into software using design best practices.
I am a computer science graduate student. I am broadly interested in Machine Learning and NLP. I have worked on text simplification, information extraction, machine translation, and knowledge discovery research in the past. I am interested in commonsense knowledge reasoning as well. I want to explore these subdomains further. I want to work on NLP research that helps in achieving Artificial General Intelligence.
I am a 5th-year PhD student at UMass Amherst. I work on various research topics about neural retrieval, such as weak supervision, multi-modal retrieval, and cut-off prediction, under the supervision of Prof. Bruce Croft and Prof. Hamed Zamani. My recent interest is in using the prediction of weakly supervised models or unsupervised generative models to improve the retrieval performance without adding ground truth.
George is a senior at University of Massachusetts Amherst majoring in Computer Science and Statistics/Data Science with his main research interests lying in deep learning, NLP, CV, IR, and optimization. He has previously worked on facial recognition bias and organization team modeling research and has been working with multimodal models. He is currently a member of the Information Fusion lab working at the intersection of NLP and CV. In the future, he hopes to pursue a PhD in machine learning. Outside of academics, George enjoys playing ultimate frisbee, table tennis, as well as story-driven games and exploring different cuisines.
I am a senior computer science student at UMass Amherst. I am also an international student from China. I am interested in information retrieval area and will write my honors thesis based on this Amazon challenge. I am passionate about computer technology, and I have taken several courses like CS383, CS514, and CS575. I program in Java and Python.
I am a Computer Science Master’s student at UMass Amherst. I graduated from UIUC with a degree in Math and Computer Science on 12/19. My research interests include Natural Language Processing, Conversational Information Seeking and Machine Learning.
I’m a Machine Learning researcher who focuses on practical applications in Natural Language Processing: Information Retrieval, Information Extraction, Question Answering, Knowledge Graphs, Recommendation Systems. I have solved challenging problems during my work at 3 AI start-ups (adtech, fintech, biotech) and in academia (RA at National University of Singapore).
Hamed Zamani - Faculty advisor