Most linked-to pages
Showing below up to 222 results in range #21 to #242.
View (previous 500 | next 500) (20 | 50 | 100 | 250 | 500)
- Spaced writing inbox (7 links)
- Iterated amplification (7 links)
- Orbit (7 links)
- Anki deck options (7 links)
- Big card (7 links)
- AI timelines (7 links)
- Simple core of consequentialist reasoning (7 links)
- Continuous takeoff (7 links)
- Hard takeoff (6 links)
- List of techniques for making small cards (6 links)
- Emotional difficulties of spaced repetition (6 links)
- List of disagreements in AI safety (6 links)
- AlphaGo (6 links)
- Duolingo (6 links)
- MIRI (6 links)
- AI safety community (6 links)
- Duolingo for math (6 links)
- Braid for math (6 links)
- Missing gear for intelligence (6 links)
- Narrow window argument against continuous takeoff (6 links)
- Will there be significant changes to the world prior to some critical AI capability threshold being reached? (5 links)
- Spaced repetition as soft alarm clock (5 links)
- Interaction reversal between knowledge-to-be-memorized and ideas-to-be-developed (5 links)
- Tao Analysis Flashcards (5 links)
- Explanation science (5 links)
- Video games comparison to math (5 links)
- Counterfactual of dropping a seed AI into a world without other capable AI (5 links)
- Braid (5 links)
- Spaced proof review as a way to invent novel proofs (5 links)
- The Witness (5 links)
- Make new cards when you get stuck (5 links)
- Something like realism about rationality (5 links)
- Tinkering in math requires loading the situation into working memory (5 links)
- Incremental reading in Anki (4 links)
- Explorable explanation (4 links)
- Human safety problem (4 links)
- Prosaic AI (4 links)
- Integration card (4 links)
- Make Anki cards based on feedback you receive (4 links)
- Card sharing (4 links)
- Spaced repetition experiments take months to complete (4 links)
- Spaced everything (4 links)
- There is room for something like RAISE (4 links)
- Dealing with bad problems in spaced proof review (4 links)
- AI safety (4 links)
- Coherence and goal-directed agency discussion (4 links)
- Anki deck philosophy (4 links)
- Daniel Kokotajlo (4 links)
- Proof card (4 links)
- HRAD (4 links)
- Thinking Mathematics (4 links)
- Missing gear vs secret sauce (4 links)
- Richard Ngo (4 links)
- List of terms used to describe the intelligence of an agent (3 links)
- Cloze deletion (3 links)
- Graduating interval (3 links)
- AI prepping (3 links)
- Optimization daemon (3 links)
- Spaced inbox (3 links)
- Resource overhang (3 links)
- Learning-complete (3 links)
- Iteration cadence for spaced repetition experiments (3 links)
- Corrigibility (3 links)
- Online question-answering services are unreliable (3 links)
- HCH (3 links)
- Architecture (3 links)
- Explosive aftermath (3 links)
- Linked list proof card (3 links)
- Carl Shulman (3 links)
- Spaced proof review routine (3 links)
- List of success criteria for HRAD work (3 links)
- Mapping mental motions to parts of a spaced repetition algorithm (3 links)
- IDA (3 links)
- Spaced proof review as a way to understand key insights in a proof (3 links)
- Evolution (3 links)
- Choosing problems for spaced proof review (3 links)
- Buck (3 links)
- Difficulty of AI alignment (3 links)
- Learner (3 links)
- AI Impacts (3 links)
- Finiteness assumption in explorable media (3 links)
- Ambitious value learning (3 links)
- My beginner incremental reading questions (3 links)
- Value learning (3 links)
- Scenius (3 links)
- Vow of silence (3 links)
- Narrow vs broad cognitive augmentation (3 links)
- The Secret of Psalm 46 (3 links)
- Intelligence amplification (3 links)
- Lumpiness (3 links)
- Deck options for proof cards (3 links)
- Instrumental convergence (3 links)
- Comparison of terms related to agency (3 links)
- Goalpost for usefulness of HRAD work (3 links)
- Sudden emergence (3 links)
- Booster card (3 links)
- Hamish Todd (3 links)
- Large graduating interval as substitute for putting effort into making atomic cards (3 links)
- Spaced proof review is not about memorizing proofs (3 links)
- Do an empty review of proof cards immediately after adding to prevent backlog (3 links)
- Decisive strategic advantage (3 links)
- Rapid capability gain (3 links)
- Orthogonality thesis (3 links)
- Spaced repetition constantly reminds one of inadequacies (3 links)
- Empty review (3 links)
- Tim Gowers (3 links)
- Ben Garfinkel (3 links)
- Changing selection pressures argument (3 links)
- FOOM (3 links)
- Updateless decision theory (2 links)
- Pascal's mugging and AI safety (2 links)
- Spaced repetition allows graceful deprecation of experiments (2 links)
- Progress in self-improvement (2 links)
- My take on RAISE (2 links)
- Hardware overhang (2 links)
- Soft-hard takeoff (2 links)
- One wrong number problem (2 links)
- Dual ratings for spaced inbox (2 links)
- Dario Amodei (2 links)
- Spaced repetition world (2 links)
- Spaced repetition and cleaning one's room (2 links)
- Word explanations are already great (2 links)
- Asymmetric institution (2 links)
- Video games allow immediate exploration (2 links)
- Optimal unlocking mechanism for booster cards is unclear (2 links)
- Video game (2 links)
- The Hour I First Believed (2 links)
- Probability and statistics as fields with an exploratory medium (2 links)
- Use temporary separate Anki decks to learn new cards based on priority (2 links)
- Nate (2 links)
- Analyzing disagreements (2 links)
- Act-based agent (2 links)
- Ongoing friendship and collaboration is important (2 links)
- Using spaced repetition to improve public discourse (2 links)
- Kasparov window (2 links)
- What makes a word explanation good? (2 links)
- Debate (2 links)
- Tutoring heuristic for prompt writing (2 links)
- List of reasons something isn't popular or successful (2 links)
- Self-graded prompts made for others must provide guidance for grading (2 links)
- Should booster cards be marked as new? (2 links)
- Representing impossibilities (2 links)
- Parasitizing on popularity (2 links)
- Creative forgetting (2 links)
- Can spaced repetition interfere with internal sense of relevance? (2 links)
- Rob Bensinger (2 links)
- Nate Soares (2 links)
- Asymmetry of risks (2 links)
- RAISE (2 links)
- Giving advice in response to generic questions is difficult but important (2 links)
- There is pressure to rush into a technical agenda (2 links)
- People watching (2 links)
- Daniel Dewey (2 links)
- Fractional progress argument for AI timelines (2 links)
- Working memory (2 links)
- On Classic Arguments for AI Discontinuities (2 links)
- Discontinuities in usefulness of whole brain emulation technology (2 links)
- Equivalence classes of prompts (2 links)
- Add all permutations of a card to prevent pattern-matching (2 links)
- 3blue1brown (2 links)
- One-sentence summary card (2 links)
- OpenAI (2 links)
- AGI (2 links)
- The mathematics community has no clear standards for what a mathematician should know (2 links)
- Content (2 links)
- Hardware-driven vs software-driven progress (2 links)
- Solomonoff induction (2 links)
- It is difficult to find people to bounce ideas off of (2 links)
- Rohin (2 links)
- Recursive self-improvement (2 links)
- Hnous (2 links)
- Feeling like a perpetual student in a subject due to spaced repetition (2 links)
- Expert response heuristic for prompt writing (2 links)
- Hnous927 (2 links)
- Gwern (2 links)
- Informed oversight (2 links)
- Combinatorial explosion in math (2 links)
- Use paper during spaced repetition reviews (2 links)
- Asynchronous support (2 links)
- It is difficult to get feedback on published work (2 links)
- Importance of knowing about AI timelines (2 links)
- Busy life periods and spaced inbox (2 links)
- Goal-directed (2 links)
- Duolingo does repetition at the lesson level (2 links)
- SuperMemo shortcuts (2 links)
- Piotr Wozniak (2 links)
- Vipul (2 links)
- Maintaining habits is hard, and spaced repetition is a habit (2 links)
- Permutation trick (2 links)
- Open Philanthropy (2 links)
- Secret sauce for intelligence vs specialization in intelligence (2 links)
- Belief propagation (2 links)
- Unreliability of online question-answering services makes it emotionally taxing to write up questions (2 links)
- AlphaGo as evidence of discontinuous takeoff (2 links)
- Mass shift to technical AI safety research is suspicious (2 links)
- Amplification (2 links)
- Factored cognition (2 links)
- MTAIR project (2 links)
- Mesa-optimization (2 links)
- Slow takeoff (2 links)
- Spaced repetition systems remind you when you are beginning to forget something (2 links)
- Potpourri hypothesis for math education (2 links)
- Category:Spaced repetition (2 links)
- Mesa-optimizer (2 links)
- Abram (2 links)
- Unbounded working memory assumption in explanations (2 links)
- Superintelligence (2 links)
- Different mental representations of mathematical objects is a blocker for an exploratory medium of math (2 links)
- Pascal's mugging (2 links)
- Central node trick for remembering equivalent properties (2 links)
- Reverse side card for everything (2 links)
- List of arguments against working on AI safety (2 links)
- Reward engineering (2 links)
- Eric Drexler (2 links)
- Good and Real (2 links)
- AI safety is not a community (2 links)
- Single-architecture generality (2 links)
- Narrow value learning (2 links)
- Content sharing between AIs (2 links)
- Snoozing epicycle (2 links)
- Managing micro-movements in learning (2 links)
- We still don't know how to systematically write great word explanations (2 links)
