Bootstrapping
One word, taught differently in 5 fields. Each use is given apart.
In AI & ML
Updating value estimates based on other estimated values rather than final actual outcomes.
In Theory & Algos
An expensive operation in homomorphic encryption used to reduce noise in ciphertexts after multiple operations.
In Systems & CS
A system of augmenting human intelligence by co-transforming or co-evolving humans and the machines they use to solve complex problems.
In Mathematics
An analytical technique used to improve an estimate of regularity by showing that if a solution has a certain order of smoothness, its derivatives possess a similar or better order.
In Physics
The game of trying to solve or constrain physical theories using general principles like symmetries, relativity, and unitarity without necessarily relying on a Lagrangian or microscopic description.
Know first
Named as a prerequisite by the lectures that define this. The number is how many say so. A grey one has no page yet.
Opens up
Concepts that name this one as a prerequisite.
In AI & ML
3Each in their own terms, from the easiest level.
David Silver
Updating estimates based on other estimates, which typically improves sample efficiency in reinforcement learning.Yaser Abu-Mostafa
Sampling randomly from your data set in order to get different data sets.
In Theory & Algos
1In Systems & CS
1In Mathematics
1In Physics
1Lectures that assume it
5Connections
21The lectures that define it or assume it, and the concepts next to it.
Concepts8
Show 2 moreShow fewer
Lectures13
- Value Function Approximation & Deep Q-Networks (DQN)
- A Report on the Navier-Stokes Problem
- Bias-Variance Tradeoff
- Model-Free Prediction — Monte Carlo, TD(0) & TD(λ)
- Reality Labs Research: A Look Back, A Look Forward
- Lorentzian Conformal Field Theory - Part 1
Show 7 moreShow fewer
- Endpoint Replay: Compressing the Recency Buffer in Deep Reinforcement Learning
- Cryptographic Proofs, Lattices & Hardness Assumptions: A Polylogue
- Model-Free Control — ε-Greedy SARSA & Off-Policy Q-Learning
- Trust Region Policy Optimization (TRPO) & Monotonic Policy Improvement
- Integrating Learning and Planning
- Exploration of deep reinforcement learning algorithms
- Exploration of advanced topics in deep reinforcement learning