### Balancing learning v/s contributing >Do research on topics you' have a good handle to make original contribution, and continue learning to understand what others are uncovering about reality I love learning so my bias is generally towards reading books, papers, doing podcasts, listening to lectures, etc. While learning is important, I've felt a distinct lack of creative problem solving in my work. I'm craving for making an original contribution to the scientific community -- no matter how small it is to begin with. There's so much to read that I must have a general direction of exploration and only read within that direction. Moreover, **should block time to do original thinking and time to do learning.** ### Choosing problems to attack Choose problems / directions where expected impact is large, which are typically at the **intersection of two relatively new fields (say qualia and quantum information theory) that you are comfortable with**. Here's why: - Not many researchers working in it. Less competition. - New fields mean not enough established results, so continual progress possible - Provides for fresh way of looking, so possibility of major breakthroughs Avoid **working in existing fields** (QM, GR, QFT) as many brilliant minds have worked on for decades and it's likely that now in these fields: - Only incremental progress is possible - The groundwork needed to solve even those incremental problems would be enormous Generally, there's an inverse relationship between how interesting a problem is and how tractable it is. Natural attraction is to attack big problems (like quantum gravity) and many scientists are doing it, but they're incredibly intractable. However, there are problems that can be easily solved (like documenting genome of a random new species) but they're not so interesting in itself. **Key is to strike the right balance between interestingness and tractability.** Also, solving questions is an easier activity as compared to asking important ones. So **focus on framing the right questions first** and only then begin answering them. ### Directions I'm excited by: **As of 7 Oct 2021** - Quantifiable models of qualia / consciousness - Consequences of taking anthropic principle seriously - Foundations of Quantum Mechanics - Link between information and physics - Origin of space and time ### Research roadmap **Research** - Some aspect of qualia / consciousness **Learning** - Quantum Field Theory (via no BS book) - General relativity (via extremely gentle introduction book) ### Archive of Directions I was excited by **As of March 2019** - neural correlates of consciousness - Animal consciousness - Evolution of brain - Insect minds - Loss of consciousness / coma - how do we get conscious access to very specific info - like we close our eyes and know how far is the wall behind us, that’s very specific - why are we able to get conscious access of certain kinds of information but not for others **As of Nov 2018** Focus for next 1 year until 2020 - Algorithmic information dynamics - Interpretation of deep learning - Proof-of-human / truly universal basic income Understanding nature of intelligence and consciousness by making artificial conscious agents. **Previously as of 27 April 2018** - algorithmic information theory - Complex adaptive systems - Evolutionary systems - Neurocomputaton and Neuroevolution - Economy-culture-tech loop - Linguistics