Skip to content

Latest commit

 

History

History
231 lines (148 loc) · 3.88 KB

timelog.md

File metadata and controls

231 lines (148 loc) · 3.88 KB

Timelog

  • Using deep reinforcement learning to solve the multiple sequence alignment problem
  • Joachim Vanneste
  • 2460800v
  • Kevin Bryson

Week 1

25 Sep 2023

  • 2 hours Research into MSA, progressive alignment and optimisation techniques

26 Sep 2023

  • 0.5 hours Set-up git repository
  • 3 hours Research into iterative alignment and probabilistic methods

27 Sep 2023

  • 1 hour Plan for project
  • 2 hours Plan for meeting

28 Sep 2023

  • 1 hour Meeting

29 Sep 2023

30 Sep 2023

01 Oct 2023

  • 4 hours Reseach for literature review

Week 2

02 Oct 2023

  • 5 hours Reseach for literature review

03 Oct 2023

  • 5 hours Start writing literature review

04 Oct 2023

  • 2 hours Start of alignment agent

05 Oct 2023

  • 5 hours DQN testing with connect 4

06 Oct 2023

07 Oct 2023

  • 3 hours CNN to extract features

08 Oct 2023

Week 3

09 Oct 2023

  • 3 hours DQN finished and training loop started

10 Oct 2023

  • 1 hour Training loop finished

11 Oct 2023

  • 2 hours Meeting preparation

12 Oct 2023

  • 1 hour Meeting

13 Oct 2023

  • 3 hours Literature review
  • 1 hour Meeting write-up

14 Oct 2023

15 Oct 2023

  • 8 hours Work on MSA RL agent - gradient updates

Week 4

16 Oct 2023

  • 3 hours Finish RL agent - slighly changed DQN model

17 Oct 2023

18 Oct 2023

19 Oct 2023

  • 3 hours Start RL agent training

20 Oct 2023

  • 5 hours Look into saving weights for transfer learning

21 Oct 2023

  • 2 hours First metrics from trained agent on 5x5s

22 Oct 2023

  • 1 hour RL agent hyperparameter optimisation

Week 5

23 Oct 2023

  • 3 hours Look into MARL, cooperative AI

24 Oct 2023

  • 3 hours More research into MARL
  • 2 hours Work on literature review

25 Oct 2023

  • 5 hours Work on literature review MSA section

26 Oct 2023

  • 3 hours Literature review
  • 1 hour Meeting write-up

27 Oct 2023

28 Oct 2023

29 Oct 2023

  • 3 hours Collect more data
  • 3 hours Test with larger sequences

Week 6

30 Oct 2023

31 Oct 2023

01 Nov 2023

  • 5 hours Work on literature review DQN section

02 Nov 2023

  • 2 hours Work on literature review
  • 2 hours Meeting
  • 1 hour Meeting write-up

01 Nov 2023

  • 5 hours Look into creating a new layer

02 Nov 2023

03 Nov 2023

04 Nov 2023

  • 2 hours Start work on new layer - created pairwise kernel

05 Nov 2023

  • 3 hours Work on literature review

Week 7

06 Nov 2023

  • 2 hours Reseach into real data sets to use

07 Nov 2023

  • 2 hours New network created with new layer
  • 1 hour Think about whether to use FC layers

07 Nov 2023

08 Nov 2023

  • 4 hours Work on interm report

09 Nov 2023

  • 2 hours Work on literature review
  • 2 hours Meeting
  • 1 hour Meeting write-up

10 Nov 2023

  • 2 hours Work on interm report

11 Nov 2023

12 Nov 2023

Week 8

13 Nov 2023

  • 2 hours Reseach into RL for MSA for report

14 Nov 2023

  • 6 hours Work on interm report

15 Nov 2023

  • 2 hours Collect histograms
  • 2 hours Running old model

16 Nov 2023

  • 4 hours Work on interm report

17 Nov 2023

  • 2 hours Work on literature review
  • 2 hours Meeting
  • 1 hour Meeting write-up

18 Nov 2023

19 Nov 2023

Week 11

15 Jan 2024

  • 2 hours Complexity analysis of my method

16 Jan 2024

  • 2 hours Complexity analysis of several MSA methods to compare

17 Jan 2024

18 Jan 2024

19 Jan 2024

  • 3 hours Planning MARL solution

20 Jan 2024

  • 4 hours Implementing start of MARL - black and white agents created

21 Jan 2024

Week 12

22 Jan 2024

  • 2 hours Implementing of MARL solution

23 Jan 2024

  • 1 hour Complexity analysis resutls plotted

24 Jan 2024

25 Jan 2024

26 Jan 2024

  • 2 hour Meeting and write up