2019 | October | Joined EA hotel on October 23rd. Attended the Learning-by-doing AI safety workshop. |
Nov-Dec | Studied and read papers about AI safety with focus on wireheading and reward hacking. Increased my background knowledge about computational complexity and algorithms; improved my mathematical problem solving skills. | |
2020 | January | Research on wireheading and reward hacking. Improved my mathematical background and problem solving skills. |
February | Read Anthropic Bias and focused on decision theory. Research proposal phase of AI Safety Camp. | |
March | Research in decision theory. Research on the topic of goal-directedness for AI Safety Camp. | |
April | Research on the mathematical formalisation of goal-directedness for AI Safety Camp. | |
May | Attended AI Safety Camp and some discussions at Web-TAISU. Research on goal-directedness and decision theory. | |
June | Sent two applications to LTFF. Research on goal-directedness and decision theory. | |
July | Research in decision theory. Stayed at the hotel for only two weeks. | |
August | Wrote a short essay on decision theory. Some work on a literature review on goal-directedness. Read some of Magnus Vinding’s books, including Suffering-Focused Ethics. | |
September | Applied to FHI RSP. Various readings in AI safety. | |
October | Read Writing Science by Schimel. Wrote a research proposal on AI alignment. | |
November | Edited the research proposal. Read Life 3.0 by Tegmark, How to Create a Vegan World by Leenaert, The Intentional Stance by Dennett. Some work on a literature review on goal-directedness. | |
December | Worked on a literature review on goal-directedness. Sent the research proposal to potential funders and mentors. Applied to CHAI internship. | |
2021 | (All year long) Research on a project about goals in intelligent agents and metaethics. | |
January | Read Practical Reasoning about Final Ends by Richardson. Read the Metaethics sequence by Yudkowsky. | |
February | Read Epistemology by Feldman, Artificial Superintelligence by Yampolskiy, papers on a formal model of AGI. | |
March | Organised research ideas, various readings, applied to LTFF. |