Armando Fortes

fmq22 [at] mails [dot] tsinghua [dot] edu [dot] cn

I am a first-year Master's student in the Department of Computer Science and Technology at Tsinghua University, advised by Prof. Jun Zhu in the Tsinghua Statistical Artificial Intelligence & Learning (TSAIL) Group.

My research interests are in Machine Learning (ML) and Natural Language Processing (NLP) in general, including representation learning, reinforcement learning, knowledge graph, and graph neural networks.

     

profile photo
News
Education

Tsinghua University
Sept 2022 - July 2024

  MSc in Computer Science and Technology
  Advisor: Prof. Jun Zhu

Tsinghua University
Sept 2021 - July 2022

  Graduate Visiting Program in Department of Computer Science and Technology

Instituto Superior Técnico, University of Lisbon
Sept 2018 - July 2021

  BSc in Computer Science and Enginnering


Selected Projects
XRDict: XLM-R Cross-Lingual Reverse Dictionary

NLP Research Project @ Tsinghua University.

The reverse dictionary is a tool which allows finding the words that best represent a given description/concept. We propose a cross-lingual solution based on the pretrained multilingual language model XLM-RoBERTa.

[Code] [Report coming soon...] [Poster coming soon...]
Image Matching Challenge

CVPR 2022 | Kaggle Competition Silver Medal 🥈 (34th out of 642 teams).

Finding point-to-point correspondences between images, a crucial task for several 3D Computer Vision problems, such as Structure-from-Motion (SfM), Simultaneous Localization and Mapping (SLAM), and Visual Localization.

[Code] [Report] [Presentation]
TMALL Repeat Buyers Prediction

Alibaba | Tianchi Competition Top 0.7% solution 🏆 (51st out of 6803 teams).

Merchants run big promotions on particular dates to attract new buyers, however, there are many one-time deal hunters. Tmall.com made their accumulated user behaviour logs available and hosted a related competition, in order to identify who can be converted into repeated buyers.

[Code] [Report] [Presentation]
Petfinder.my - Pawpularity Contest

Kaggle Competition Bronze Medal 🥉 (205th out of 3537 teams).

Millions of stray animals wait for a new home and family. With the aim of understanding the impact of pet photo features on their popularity, PetFinder.my designed the Pawpularity metric and hosted a competition.

[Code] [Report] [Poster]

Website inspired from here.