EXPLORE / 06
Terms index
Specialized AI and software-development terms with concise explanations
Entity records
4520 records · 91/113
- 3601
Reversible Neural Network
A network structure allowing lossless bidirectional information propagation, often used in generative models.
Website pending - 3602
Domain Randomization
Randomizing simulation parameters to improve model generalization to reality
Website pending - 3603
Options Framework
A hierarchical decision structure in RL representing temporally abstracted actions.
Website pending - 3604
Variational Inference
An approximate Bayesian method that converts posterior probability calculation into an optimization problem.
Website pending - 3605
Zero-Sum Game
Game model where one party's gain equals the other's loss, common in competitive games
Website pending - 3606
Intrinsic Motivation
Behavior driven by internal psychological states like curiosity, rather than external rewards
Website pending - 3607
Jacobian Matrix
Matrix of first-order partial derivatives describing local linear approximation, used for computing transformation volume elements
Website pending - 3608
Normalizing Flow
A generative model that maps simple distributions to complex target distributions via invertible transformations
Website pending - 3609
Neural ODE
Treats neural networks as continuous-time dynamical systems, describing hidden state changes via differential equations
Website pending - 3610
Task Decomposition
The process of breaking down complex problems into independently solvable sub-problems.
Website pending - 3611
Iterated Amplification
A technique training AI by recursively decomposing tasks and amplifying human judgment.
Website pending - 3612
Open Source
Software licensing model that allows source code to be viewed, modified, and distributed
Website pending - 3613
Sparse Reward
An environment characteristic where feedback is provided only upon task completion or rarely, making exploration difficult.
Website pending - 3614
Intrinsic Reward
A non-environmental reward signal driven by the agent's internal state uncertainty or novelty.
Website pending - 3615
Random Network Distillation (RND)
An exploration method generating intrinsic rewards by minimizing the difference between a target random network and a predictor network.
Website pending - 3616
Cross-Domain Transfer
Applying knowledge learned in one domain or task to a different but related domain or task.
Website pending - 3617
Energy-Based Model
A generative framework that assigns an energy score to data points rather than explicitly modeling probability distributions.
Website pending - 3618
Offline Learning
Method of training policies using historical data without real-time environment interaction
Website pending - 3619
Model-Based RL
Reinforcement learning approach that learns environment dynamics to aid decision-making and planning
Website pending - 3620
Batch Size
The number of samples used in one iteration of parameter update.
Website pending - 3621
Gradient Noise Scale
A statistical measure quantifying the difference between mini-batch gradient estimates and full-data gradients.
Website pending - 3622
Long-term AI Safety
Research field focusing on existential risks posed by future superintelligence
Website pending - 3623
Computational Complexity
A theoretical measure of resources (time, space) required to solve a specific problem
Website pending - 3624
Robust Classification
The ability of a model to maintain accurate predictions despite input perturbations or noise
Website pending - 3625
Exploration
The process where an agent tries new actions to discover unknown states or rewards.
Website pending - 3626
MARL
A method where multiple agents learn policies via trial-and-error in a shared environment.
Website pending - 3627
Neuron Interaction
The mechanism of signal transmission and collaboration between nodes in a neural network
Website pending - 3628
Capped-Profit Company
A corporate form that allows profit-making but caps returns or directs surplus to social good
Website pending - 3629
Mode Coverage
The ability of a generative model to cover all major modes in the data distribution.
Website pending - 3630
Energy-Based Model (EBM)
A generative model that defines a data distribution via an energy function.
Website pending - 3631
Next Token Prediction
Task of predicting the probability distribution of the next discrete unit based on preceding sequences.
Website pending - 3632
Adversarial Perturbation
Carefully crafted small noise added to inputs to mislead model outputs
Website pending - 3633
Generative Modeling
Machine learning method that learns data distributions to generate new samples.
Website pending - 3634
Sparse Attention
A mechanism that computes attention only between selected positions to reduce complexity.
Website pending - 3635
Collective Action Problem
A situation where individual rational choices lead to suboptimal group outcomes.
Website pending - 3636
Supercomputing
High-performance computing systems for massive complex tasks
Website pending - 3637
Co-adaptation
The process where multiple entities mutually adjust their behaviors during interaction to optimize performance
Website pending - 3638
Emergence
Properties or behaviors arising from complex systems that are not present in individual components
Website pending - 3639
Adversarial Attack
Techniques that use small perturbations to mislead ML models into making incorrect predictions.
Website pending - 3640
Sim-to-Real Transfer
The process of applying models trained in computer simulations directly to robots in the physical world.
Website pending