Tags
- 09
- 0.999...
- 2010s
- 3D
- 6
- A
- Abbreviation
- Absolute
- Accelerated
- Acceleration
- Achievable
- Acro
- Across
- Actual
- Adam Wa
- Adaptation
- Adaptive Behavior
- Adaptive filter
- Adaptive learning
- Algorithm
- Algorithmic efficiency
- Almost
- Alpha
- An
- Analogy
- Analysis
- And 1
- Anita Lo
- Application
- Approaches
- Appropriation
- Approximation
- Arise
- Arithmetic mean
- Artificial
- Artificial neural network
- Assumption
- As Was
- Backpropagation
- Backtracking
- Backtracking line search
- Balancing
- Ball
- Based on
- Basic
- Batch
- Behavior
- Behind
- Being
- Benefit
- Benefits
- Beta
- Bisection
- Bisection method
- Boris Polyak
- Both Sides
- Bound
- Bounds
- Branches
- Broyden–Fletcher–Goldfarb–Shanno algorithm
- Burden
- Calculated
- Cause
- Choice
- Class
- Classic
- Classical
- Class of
- Closed form
- Closed-form expression
- Code
- Combination
- Combined
- Combining
- Community
- Comparison
- Compete
- Compromise
- Computation
- Compute!
- Computer
- Computer Science
- Computer scientist
- Computing
- Concept
- Consequence
- Constant
- Contemporary
- Context
- Contrast
- Contribution
- Converge
- Convergence
- Convex
- Convex optimization
- Coursera
- Cycles
- Data
- Data set
- David Rumelhart
- Decay
- Decrease
- Decreasing function
- De Facto
- De facto standard
- Delta
- Demonstration
- Denotation
- Dependent and independent variables
- Derivative
- Descent
- Described
- Determine
- Diagonal
- Differences
- Differentiable function
- Dimension
- Direction
- Divergence
- Divide
- Division
- Division by zero
- Doctor of Philosophy
- D.O.E.
- Domination
- Dynamics
- Economic system
- Efficient
- Eigenvalues and eigenvectors
- Elements
- Empirical
- Empirical risk minimization
- Enabling
- Enjoy
- Enormous
- Epsilon
- Equation
- Equivalent
- Establishment
- Estimate
- Estimating equations
- Estimation
- Estimator
- Evaluation
- Evaluations
- Every
- Every Step
- Exchange
- Expense
- Exponential
- Exponential decay
- Exponential family
- Extreme
- Fact
- Factor
- Familie
- Families
- Fashion
- Filter
- Fine Tuning
- Finite difference
- Fit
- Fixed
- Following
- Follows
- Forgetting
- Formula
- Fraction
- Frank Rosenblatt
- Frequentist inference
- Full waveform inversion
- Functional
- Fx Ya
- Gamma
- Generalization
- Generalized linear model
- Geoffrey Hinton
- Geophysics
- Global optimization
- GOE
- Goes
- Gradient
- Gradient descent
- Gradient method
- Graph
- Graphics
- Guarantee
- Guidance
- Hand
- Harmonic oscillator
- Heavy
- Herbert Robbin
- Herbert Robbins
- Hidden
- Higher Learning
- Hinton
- Hyperparameter
- If
- Ilya Sutskever
- Implement
- Implementation
- Implicit
- IN-6
- Include
- Incomplete
- Index
- Indication
- Induce
- Initial
- Inner product space
- Innervisions
- In Practice
- Inspired
- Instability
- Intercept
- Interior
- Interpretation
- Interpretations
- Introduction
- Invented
- Inversion
- Iteration
- Iterative method
- Jack Kiefer
- Jacob Wolfowitz
- James Marten
- James Martens
- Keep
- KEY Difference
- K-means clustering
- Known
- Langevin
- Langevin dynamics
- Language
- Language processing in the brain
- Large-scale machine learning
- Last Alliance
- Layers
- Learning
- Learning rate
- Least mean squares filter
- Least squares
- Less
- Let
- Libraries
- Likelihood function
- Limited-memory BFGS
- Linear
- Linear combination
- Linear model
- Linear regression
- Line level
- Line search
- Link
- LMS
- Logistic
- Logistic function
- Logistic regression
- Loop
- Loss
- Loss function
- Machine
- Machine learning
- Machines
- Magnitude
- Manner
- Martens
- Mathematical optimization
- Mathematician
- Matrix
- Maxima and minima
- Maximum
- Maximum likelihood estimation
- McQueen
- Mean squared error
- M-estimator
- Method
- Methods
- Mini-batch
- Mini-batch gradient descent
- Minimisation
- Minimum
- Model
- Models
- Modified
- Moment
- Moments
- Momentum
- Monro
- Most
- Moving average
- Multiplication
- Natural
- Natural language
- Natural language processing
- Need
- Nesterov
- Nesterov accelerated gradient
- Network
- Neural
- Neural network
- Neural Networks
- Newton County John Does
- Next
- No
- Normalization
- Note
- Number
- Numerical
- Numerical analysis
- Numerical integration
- Numerical stability
- Observation
- Observations
- Old
- One Year
- One Year Later
- Online and offline
- Only
- Opposed
- Optimization
- Optimization problem
- Optimization problems
- Optimization techniques
- Optimize
- Ordinary
- Originally
- Oscillations
- Outer
- Outer product
- Over
- Pâ
- Paper
- Parameter
- Parameters
- Particle
- Particular
- Pass
- Pavement
- Perceptron
- Perform
- Performance
- Perspective
- PHD
- Physics
- Place
- Poisson
- Poisson regression
- Polyak
- Popular
- Popularity
- Pragmatism
- Predecessor
- Prediction
- Presenting
- Prevention
- Principle
- Probability
- Problematic
- Procedure
- Proceed
- Processing
- Product
- Projection
- Propagation
- Properties
- Property
- Proportionality
- Proposals
- Proximal gradient method
- Proximal gradient methods for learning
- Pseudocode
- Pseudoconvexity
- Publishing
- Randomness
- Range
- Rate schedule
- Recognized
- Record
- Reduce
- Reduction
- Regression
- Regression models
- Remains
- Replace
- Replacement
- Require
- Research
- Researcher
- Respect
- Restrictiveness
- Result
- Return to Cookie Mountain
- Revealed
- Risk
- Rmsprop
- Robbins
- Root Mean
- Root mean square
- Rosenblatt
- Rprop
- Ruppert
- Saint Laurent Boulevard
- Same Direction
- Sample
- Sampling
- Scalar
- Scaling
- Scaling factor
- Schedule
- Schedules
- Scientist
- Score
- Score function
- Search
- Second
- Second moment
- Second-order
- Selected
- Sensitive
- Set
- Settings
- SGD
- Short
- Shortened
- Shown
- Sides
- Simple extension
- Simplification
- Simulation
- Single
- Slow
- Small batches
- Smoother
- Smoothness
- So-called
- Solution
- Solve
- Solved
- Some
- Sometimes
- Sourced
- Soviet
- Sparse
- Specific
- Square
- Stable
- Stationary
- Stationary point
- Statistics
- Stem
- Steps
- Still
- Stochastic
- Stochastic approximation
- Stochastic gradient descent
- Store
- Storing
- Straight
- Strategy
- Strong
- Student
- Subsequent
- Subset
- Substitution reaction
- Sum 41
- Summation
- Sum of squares
- Support vector machine
- Supposition theory
- Sutton Monro
- Sweep
- Sweeps
- Techniques
- TensorFlow
- Term
- Terms
- Test set
- The Algorithm
- The Equation
- The first
- The Formula
- The general
- The gradient
- The Loop
- The Method
- Then
- The Norm
- The Objective
- Theory
- The other
- The place
- The Procedure
- The Score
- The time
- The way
- The Weight
- The work
- Today
- Torch
- Total
- Training
- Traveling
- Trial
- Uniform distribution
- Uniform norm
- Unknown
- Unstable
- Updates
- URL shortening
- Uses
- Variable
- Variant
- Vectorization
- Virtually
- Vowpal Wabbit
- W
- Wabbit
- Waveform
- Weight
- Weight transfer
- We-Wish
- When
- Wide
- Wikipedia
- Williams
- Wish
- With high probability
- Wolfowitz
- Yurii Nesterov
- Zero
- Zeros