Index
A
- addressing mechanisms, Neural Turing Machine (NTM)
- content-based addressing / Content-based addressing, Copy tasks using NTM
- location-based addressing / Location-based addressing, Copy tasks using NTM
- Adversarial Meta Learning (ADML)
- about / Adversarial meta learning, ADML
- Fast Gradient Sign Method (FGSM) / FGSM
- building / Building ADML from scratch
- data points generation / Generating data points
- FGSM, using / FGSM
- single layer neural network / Single layer neural network
- implementing / Adversarial meta learning
- Artificial General intelligence (AGI) / Algorithm
- audio recognition model
- building, siamese networks used / Building an audio recognition model using siamese networks
C
- CACTUS
- about / CACTUs
- task generation / Task generation using CACTUs
- CAML algorithm / CAML algorithm
- classification
- performing, prototypical networks used / Performing classification using prototypical networks
- components, Neural Turing Machine (NTM)
- controller / NTM
- memory / NTM
- read head / NTM
- write head / NTM
- concept discrimination loss / Concept discrimination loss
- concept discriminator / Concept discriminator
- concept generator / Learning to learn in concept space, Concept generator
- concept space
- about / Learning to learn in concept space
- algorithm / Algorithm
- Context Adaptation for Meta Learning (CAML)
- about / CAML
- context parameter / CAML
- shared parameter / CAML
- algorithm / CAML algorithm
- Convolutional Neural Network (CNN) / What are siamese networks?
E
- embedding functions, matching networks
- support set embedding function (g) / The support set embedding function (g)
- query set embedding function (f) / The query set embedding function (f)
- embeddings / Prototypical networks
- episodic fashion / Meta learning and few-shot
F
- face recognition
- with siamese networks / Face recognition using siamese networks
- Fast Gradient Sign Method (FGSM) / FGSM
- few-shot learning
- about / Meta learning and few-shot
- optimization, as model / Optimization as a model for few-shot learning
G
- Gaussian prototypical network
- about / Gaussian prototypical network
- radius component / Gaussian prototypical network
- diagonal component / Gaussian prototypical network
- algorithm / Algorithm
- generalized entropy measures / Theil index
- gradient agreement
- as optimization / Gradient agreement as an optimization
- weight calculation / Weight calculation
- gradient agreement algorithm
- about / Algorithm
- building, with MAML / Building gradient agreement algorithm with MAML
- data points, generating / Generating data points
- single layer neural network / Single layer neural network
- in MAML / Gradient agreement in MAML
I
- inequality measures, TAML
- about / Inequality measures
- Gini coefficient / Gini coefficient
- Theil index / Theil index
- variance of algorithms / Variance of algorithms
K
- k-shot learning / Meta learning and few-shot
L
- location-based addressing, Neural Turing Machine (NTM)
- interpolation / Interpolation, Copy tasks using NTM
- convolution shift / Convolution shift, Copy tasks using NTM
- sharpening / Sharpening, Copy tasks using NTM
M
- matching networks
- about / Matching networks
- embedding functions / Embedding functions
- architecture / The architecture of matching networks
- in TensorFlow / Matching networks in TensorFlow
- Mean Squared Error (MSE) / Loss function
- Memory-Augmented Neural Networks (MANN)
- about / Memory-augmented neural networks (MANN)
- read operation / Read operation
- write operation / Write operation
- Meta-SGD
- about / Meta-SGD
- for supervised learning / Meta-SGD for supervised learning
- building / Building Meta-SGD from scratch
- data points generation / Generating data points
- single layer neural network / Single layer neural network
- implementing / Meta-SGD
- in reinforcement learning / Meta-SGD for reinforcement learning
- Meta Imitation Learning (MIL)
- about / Meta imitation learning
- algorithm / MIL algorithm
- meta learner / Meta learner
- meta learning
- about / Meta learning
- types / Types of meta learning
- metric-based meta learning / Learning the metric space
- initializations-based meta learning / Learning the initializations
- optimizer-based meta learning / Learning the optimizer
- meta learning algorithms / Learning to learn gradient descent by gradient descent
- meta learning loss / Meta learning loss
- Model Agnostic Meta Learning (MAML)
- about / MAML
- algorithm / MAML algorithm
- in supervised learning / MAML in supervised learning
- building / Building MAML from scratch
- data points generation / Generate data points
- single layer neural networks / Single layer neural network
- implementing / Training using MAML
- in reinforcement learning / MAML in reinforcement learning
N
- Neural Turing Machine (NTM)
- about / NTM
- components / NTM
- read operation / Read operation, Copy tasks using NTM
- write operation / Write operation, Copy tasks using NTM
- erase operation / Erase operation
- add operation / Add operation
- addressing mechanisms / Addressing mechanisms, Copy tasks using NTM
- tasks, copying / Copy tasks using NTM
O
- one-shot learning / Meta learning and few-shot
P
- prototypical networks
- about / Prototypical networks
- example / Prototypical networks
- flow / Prototypical networks
- algorithm / Algorithm
- used, for performing classification / Performing classification using prototypical networks
Q
- query set / Meta learning and few-shot
- query set embedding function (f) / The query set embedding function (f)
R
- rectified linear unit (ReLU) / Face recognition using siamese networks
- Recurrent Neural Network (RNN) / Learning to learn gradient descent by gradient descent
- relation networks
- about / Relation networks
- in one-shot learning / Relation networks in one-shot learning
- in few-shot learning / Relation networks in few-shot learning
- in zero-shot learning / Relation networks in zero-shot learning
- loss function / Loss function
- building, Tensorflow used / Building relation networks using TensorFlow
- Reptile algorithm
- about / Reptile, The Reptile algorithm
- sine wave regression / Sine wave regression using Reptile
- implementing / Reptile
S
- semi prototypical networks / Semi-prototypical networks
- siamese networks
- about / What are siamese networks?
- working / What are siamese networks?
- architecture / Architecture of siamese networks
- applications / Applications of siamese networks
- reference / Applications of siamese networks
- using, in face recognition / Face recognition using siamese networks
- audio recognition model, building / Building an audio recognition model using siamese networks
- sine wave regression, Reptile
- about / Sine wave regression using Reptile
- data points generation / Generating data points
- two layered neural network / Two-layered neural network
- Stochastic Gradient Descent (SGD) / Reptile
- support set / Meta learning and few-shot, Prototypical networks
- support set embedding function (g) / The support set embedding function (g)
T
- Task-agnostic meta-learning (TAML)
- about / Task agnostic meta learning (TAML)
- entropy maximization/reduction / Entropy maximization/reduction
- working / Algorithm
- inequality minimization / Inequality minimization
- implementing / Algorithm
- task generation
- with CACTUS / Task generation using CACTUs
- tasks
- copying, Neural Turing Machine (NTM) used / Copy tasks using NTM
- TensorFlow
- relation networks, building / Building relation networks using TensorFlow
- matching networks / Matching networks in TensorFlow
- Theil index / Theil index
Z
- zero-shot learning / Meta learning and few-shot