Messages
The WNNLP'23 proceedings are now published, please have a look.
Congratulations to all the authors!
The exam webpage has now been updated with information about the presentations. We copy the relevant update here for convenience:
All teams (whether PhD or MSc) should prepare a 5-minute "lighning talk" for the final workshop on June 6th at 12:15, including a few slides (as a pdf). The track chairs will present a short task overview for each track, so you don't need to spend time and space on this in your presentation. Focus on the main contributions and results. Note that the 5-minute time limit is strict due to our tight program, so be sure to not include more material than you can realistically cover.
We have a separate public (uio-)git repo for the workshop, where all authors have been given write access: https://github.uio.no/in5550/WNNLP-2023
Make sure you push your presentati...
The submission deadline for WNNLP 2023 has now closed and we are ready to start the reviewing process. Please log into EasyChair and find the papers assigned to you. If this is your first time logging in, please create an account using USERNAME@uio.no; https://easychair.org/conferences/?conf=wnnlp2023
You will find some guidelines for how to conduct the reviewing in the IN5550 git repo, see https://github.uio.no/in5550/2023/tree/main/exam/review
Note that the deadline for submitting your reviews is by the end of Thursday May 25 (by midnight, Norwegian time).
For paper submissions and reviewing for WNNLP 2023 (i.e. our home exam workshop) we will a platform called EasyChair. See the readme for the course GitHub for the login link, and use your UiO email address on the form USERNAME@uio.no when creating your account.
Hi, here's the leader-board for the third obligatory assignment. We have only two leaders, and they are getting a bonus point if they are not already at the maximum. Congratulations!
| Team | F1 score |
|---|---|
| Amir, Cornelius and Torstein | 92% |
| Martins | 92% |
We have published an additional bonus sub-task for Obligatory assignment 3. Completing it can give you up to 4 extra points. Submit together with the main task.
Hi everyone taking the IN5550 (and IN9550) course! Congratulations on getting this far!
While you are busy with the Obligatory 3, here are some updates about what's coming next:
- There will be no lectures and group sessions during the Easter (that is, next week). On April 11, we will have a regular Q&A session on Zoom, covering bias and sustainability in deep learning for NLP. The lecture videos will be published beforehand, as usual.
- On April 18, we will present topics for the Final Exam. Note that this will be an in-person event, not a Zoom meeting or a pre-recorded lecture! Plan accordingly. You will have a week to choose one of the topics.
- Right before you start with the exam, we will have a guest talk. Murhaf Fares and Emmanuelle Laponi (Fremtind) will tell you why F1 doesn't matter (and other stor...
Hi, here's the leaderboard for the second obligatory assignment. As promised, the first 3 teams will get a bonus point, congratulations!
| Team | Accuracy |
|---|---|
| Amir, Cornelius and Torstein | 66% |
| Magnus | 61% |
| Roman and Lu (shared) |
55% |
The third obligatory assignment is now online, on the Git repository. It deals with using large pre-trained BERT language models to solve the task of named entity recognition (NER) for Norwegian.
The assignment is due April 19, but feel free to start working on it. We have already covered masked language models in the last lecture.
We will provide hands-on exercises in using BERT and HuggingFace Transformers at the next group sessions.
Hi, here's the leaderboard for the first obligatory assignment. As promised, the first 3 teams will get a bonus point, congratulations! The competition was fierce this year, so don't be sad if you're not among the first 3 teams, we saw a lot of great works when grading your assignments!
| Team | Macro F1 |
|---|---|
| Victor and Magnus | 49% |
| Amir, Cornelius and Torstein | 48% |
| Lu Xing |
43% |
The second obligatory assignment is now online, on the Git repository. It deals with using pre-trained word embeddings and recurrent neural networks to solve the natural language inference (NLI) task.
The assignment is due March 15, but feel free to start working on it. We have already covered word embeddings in the last lectures, and will cover recurrent neural networks (RNNs) very soon.
We will provide extensive practice in building RNN classifiers with PyTorch at the group sessions.
Important notice: Q&A session tomorrow (February 21) will start at 13:15, not 12:15, because of the CuttingEdgeAI: Large Language Models event.
This means we will have only 45 minutes, but it should be fine. In case any questions still remain after the Q&A session, we will discuss them in the lab on Wednesday February 22.
As announced during last week's lab session, the time slot for the QA tomorrow (07.02) will be used for this week's lab, which will be held digitally on Zoom (same Zoom room as usual). The orignal time slot for the lab on Wednesday will not be used for anything, so you will have to save your questions about this week's lecture until next week's QA (14.02)
The first obligatory assignment is now online, on the Git repository. It deals with building a simple neural document classifier using bags-of-words as features.
The assignment is due February 16, but feel free to start working on it. We have already covered linear classifiers in the last lecture, and will cover training feed-forward neural networks in the next one.
We will provide extensive practice in building neural classifiers with PyTorch at the group sessions.
The first set of pre-recorded lectures is out now, under the common title of "Supervised Machine Learning: From Linear Models to Neural Networks".
It consists of four shorter sub-lectures, and you can find the videos together with the slides at the course schedule page.
Please make sure to get yourself acquainted with the videos before our next Q&A session which will take place next Tuesday, January 31, in Zoom.
Welcome to our IN5550 course which will guide you through deep learning applications to natural language processing!
The first introductory lecture this term will be held on Tuesday, January 24, at 12:15. We will go through course logistics (including routines for assignments and the final project-based exam) and motivate the now dominant use of neural architectures in Natural Language Processing (and most other sub-fields of Artificial Intelligence). The first lecture will be in-person.
Subsequent lectures (after the introductory one) will be provided to you in a pre-recorded format. Currently, the plan is that the lecture videos will be published every Friday in the second half of the day. You can watch them whenever it is more convenient to you. Each Tuesday (at the official designated time slot of 12:15), we will have an interactive Q&A session in Zoom, dedicated to t...