按方法、领域与证据探索科学。

Regularized Support Vector Machine

Regularized Support Vector Machine extends the classic SVM by explicitly controlling the trade-off between margin maximization and training error through an L1 or L2 penalty parameter. The soft-margin formulation introduced by Cortes and Vapnik in 1995 is itself a regularized model, and later L1-SVM variants additional

machine learning

Regularized Transfer Learning

Regularized Transfer Learning applies explicit penalty terms to a transfer learning pipeline to control how much a model shifts away from source-domain knowledge when adapting to a new target domain. The regularizer discourages negative transfer — the harmful carry-over of irrelevant source patterns — while preserving

deep learning

Reinforcement Learning

Reinforcement Learning (RL) is a framework in which an agent learns to make sequential decisions by interacting with an environment, receiving scalar reward signals, and updating a policy to maximise cumulative future reward. Unlike supervised learning, no labeled examples are provided; the agent discovers optimal beha

text mining

Relation Extraction

Relation extraction is a natural-language-processing task that detects and classifies the semantic relations that hold between entities mentioned in text. Building on early kernel-based methods (Zelenko and colleagues, 2003) and later neural matching approaches (Baldini Soares and colleagues, 2019), it turns free-form

particle physics

Renormalization Group Equations

Renormalization Group Equations (RGEs) describe how the coupling constants and masses of a quantum field theory evolve with energy scale. They are fundamental tools for understanding the scale dependence of physics, predicting the behavior of coupling strengths at different energies, and connecting high-energy physics

deep learning

ResNet

ResNet (Residual Network) is a deep convolutional neural network architecture introduced by Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun at CVPR 2016. By inserting shortcut (skip) connections that carry the input of a block directly to its output — defining the block's task as learning a residual correction ra

deep learning

ResNeXt

ResNeXt is a deep convolutional neural network architecture introduced by Xie, Girshick, Dollár, Tu, and He at CVPR 2017. It extends the residual network (ResNet) design by introducing a new architectural dimension called cardinality — the number of independent, parallel transformation paths within each residual block

civil engineering

Response Spectrum Analysis

Response spectrum analysis is a linear modal method for estimating earthquake-induced forces and displacements in structures. Introduced by Housner in 1941, this approach uses design spectra that represent the maximum response of single-degree-of-freedom oscillators at different natural frequencies to compute the struc

reliability engineering

Response Surface Desirability Function

Response Surface Methodology (RSM) is a set of statistical and mathematical techniques for modeling and optimizing processes with multiple inputs (factors) and outputs (responses). The Desirability Function approach, introduced by Harrington (1965) and refined by Derringer and Suich (1980), extends RSM to solve multi-r

deep learning

Restricted Boltzmann Machine

A Restricted Boltzmann Machine is a two-layer generative probabilistic model consisting of visible (observed) and hidden (latent) binary units connected by an undirected bipartite graph with no within-layer connections. Originally introduced as the 'Harmonium' by Paul Smolensky in 1986 and powerfully revived by Geoffre

text mining

Retrieval-Augmented Generation

Retrieval-Augmented Generation (RAG) is a natural-language-processing pipeline introduced by Lewis et al. in 2020 that strengthens a large language model (LLM) with evidence fetched at inference time from an external knowledge base. Instead of relying solely on what a model memorised during training, RAG first retrieve

cryptography

Return-Oriented Programming

Return-Oriented Programming (ROP) is an exploit technique that chains together short sequences of instructions (gadgets) from existing executable code to perform arbitrary computation, bypassing security defenses like code injection prevention. Introduced by Hovav Shacham in 2007, ROP exploits code reuse to execute mal

fluid dynamics

Reynolds-Averaged Navier-Stokes

The Reynolds-Averaged Navier-Stokes (RANS) equations represent a time-averaged form of the Navier-Stokes equations developed by Osborne Reynolds in 1895. This approach decomposes turbulent flow into mean and fluctuating components, enabling practical simulation of turbulent flows by modeling turbulent stresses rather t

cryptography

Ring Signature

A ring signature is a digital signature scheme allowing a member of a group (ring) to sign a message on behalf of the group without revealing the signer's identity. Proposed by Rivest, Shamir, and Tauman in 2001, ring signatures provide signer anonymity while still proving that the signature comes from one member of a

genetics

RNA Velocity

RNA velocity is a computational method that infers the future developmental state of individual cells from single-cell RNA-sequencing data. Developed by La Manno and colleagues in 2018, RNA velocity analysis measures the direction and pace of cell state transitions by analyzing the ratio of unspliced to spliced mRNA tr

bioinformatics

RNA-seq Differential Expression

RNA-seq differential expression (DE) analysis identifies genes whose transcript abundance differs significantly between two or more biological conditions — for example, treated versus control, or diseased versus healthy tissue. Starting from raw sequencing reads, the pipeline moves through alignment, count-based normal

deep learning

RoBERTa-based Classification

RoBERTa-based Classification applies the RoBERTa pre-trained transformer — trained more robustly than BERT with dynamic masking and larger batches — to text categorisation tasks by adding a lightweight classification head on top of the [CLS] token representation and fine-tuning the entire model on labelled examples. It

machine learning

Robust Active Learning

Robust Active Learning extends the standard active learning framework to handle noisy labels, adversarial perturbations, and unreliable oracles. Rather than assuming perfect labeling, it incorporates statistical or adversarial robustness guarantees into the query selection process, maintaining sample efficiency while t

simulation

Robust Agent-Based Modeling

Robust Agent-Based Modeling (Robust ABM) integrates systematic uncertainty quantification and sensitivity analysis into agent-based simulation workflows. Rather than relying on a single parameter configuration, it explores the full parameter space to identify which inputs drive model outcomes, ensuring that conclusions

machine learning

Robust Autoencoder anomaly detection

Robust Autoencoder Anomaly Detection extends the standard autoencoder framework with robustness mechanisms — such as sparse decomposition, robust loss functions, or adversarial regularisation — so that the model learns a compact representation of normal behaviour while remaining resistant to the corrupting influence of

machine learning

Robust Boosting

Robust Boosting modifies standard boosting algorithms — such as AdaBoost or gradient boosting — by replacing the default exponential or squared loss with robust loss functions (e.g., Huber, logistic, or truncated losses) or by incorporating noise-tolerance mechanisms, so that the ensemble remains accurate even when tra

machine learning

Robust Decision Tree

A Robust Decision Tree is a decision tree variant trained with modified splitting criteria or training procedures designed to reduce sensitivity to outliers, label noise, and adversarial perturbations. Rather than minimizing standard impurity measures that are strongly affected by extreme values, robust variants use st

simulation

Robust Discrete-Event Simulation

Robust Discrete-Event Simulation (Robust DES) is a simulation methodology that extends classical discrete-event simulation by explicitly incorporating uncertainty in model parameters — such as interarrival times, service durations, and resource capacities — and evaluating system performance across worst-case or distrib

machine learning

Robust Federated Learning

Robust Federated Learning extends standard federated learning with Byzantine-tolerant aggregation rules that protect the global model against malicious, corrupted, or unreliable clients. Instead of naively averaging client gradients, robust aggregation methods such as coordinate-wise median or Krum filter out harmful u

machine learning

Robust Gaussian Mixture Model

Robust Gaussian Mixture Model replaces the standard Gaussian components with heavier-tailed distributions — most commonly Student's t-distributions — or incorporates trimming and down-weighting of outliers within the EM framework. The result is a probabilistic clustering and density-estimation method that assigns genui

machine learning

Robust Gaussian Process

Robust Gaussian Process (Robust GP) extends the standard Gaussian Process framework by replacing the Gaussian noise likelihood with a heavy-tailed distribution — typically Student-t — so that outliers in the training data exert less influence on the learned function. It retains the full probabilistic, uncertainty-quant

machine learning

Robust HDBSCAN

Robust HDBSCAN (HDBSCAN*) extends the original HDBSCAN algorithm with a robust single-linkage framework that handles noise, outliers, and clusters of varying densities more reliably. Introduced by Campello et al. (2015), it converts any density-based hierarchy into a stable flat clustering while explicitly modeling noi

machine learning

Robust Isolation forest

Robust Isolation Forest extends the classic Isolation Forest anomaly detector with strategies that reduce sensitivity to data contamination, masking effects, and biased random splits. By incorporating robustness mechanisms — such as improved subsampling, re-weighting of suspicious regions, or bias-corrected splitting —

machine learning

Robust k-means

Robust k-means is a variant of classical k-means clustering designed to resist the influence of outliers. By trimming a specified fraction of the most extreme observations before computing cluster centers, it produces stable and meaningful partitions even when the data contain noise, contamination, or heavy-tailed dist

simulation

Robust Markov Model

A Robust Markov Model applies robustness principles to Markov chains by replacing single-point transition probabilities with uncertainty sets, then optimizing against the worst-case realization. Originally developed for robust Markov decision processes in operations research, it is used wherever transition rates are es

machine learning

Robust Metric Learning

Robust Metric Learning learns a Mahalanobis distance function from labeled or pairwise-constrained data while actively resisting the distortion caused by noisy labels, corrupted examples, or outliers. By replacing standard hinge or squared losses with robust alternatives and adding regularization, it produces a distanc

simulation

Robust Microsimulation

Robust Microsimulation combines individual-level (micro) simulation with systematic uncertainty analysis — typically probabilistic sensitivity analysis — to generate outputs that are robust to parameter uncertainty, model structure assumptions, and input variability. It is widely used in health technology assessment, p

simulation

Robust Multi-Objective Optimization

Robust Multi-Objective Optimization (RMOO) is a framework for finding solutions that simultaneously optimize multiple conflicting objectives while remaining insensitive to perturbations in decision variables or problem parameters. Unlike classical MOO, RMOO explicitly incorporates uncertainty into the optimization loop

machine learning

Robust One-class SVM

Robust One-Class SVM extends the classic One-Class Support Vector Machine for novelty and anomaly detection by incorporating robustness mechanisms — such as trimmed objectives, robust kernel choices, or contamination-tolerant loss functions — that reduce the influence of heavy-tailed noise or outliers present in the tr

machine learning

Robust Online Learning

Robust Online Learning extends the online learning framework — where a model updates sequentially after each observation — by incorporating robustness mechanisms that guard against corrupted labels, adversarial examples, heavy-tailed noise, and concept drift. The result is a sequential learner that maintains bounded re

machine learning

Robust Random Forest

Robust Random Forest extends the standard Random Forest ensemble by incorporating mechanisms that reduce the influence of outliers, label noise, and corrupted observations. Rather than treating all training instances equally, it applies weighting or filtering strategies so that noisy or anomalous samples contribute les

psychometrics

Robust Rasch Model

The robust Rasch model applies the standard one-parameter logistic Rasch framework with estimation procedures designed to limit the influence of outlying item responses, aberrant respondents, or mild model violations, producing stable item and person parameter estimates that are less sensitive to data contamination tha

simulation

Robust Scenario Analysis

Robust Scenario Analysis evaluates a set of candidate strategies across a structured collection of plausible future scenarios and selects the strategy that performs acceptably well — or best in the worst case — regardless of which scenario materializes. It merges scenario planning with robustness criteria such as maxim

simulation

Robust Sensitivity Analysis

Robust Sensitivity Analysis (RSA) systematically evaluates how much variation in model outputs can be attributed to uncertainty or variation in model inputs, with an explicit focus on conclusions that remain valid across a wide range of plausible input conditions. It goes beyond standard sensitivity analysis by asking

machine learning

Robust Stacking Ensemble

Robust Stacking Ensemble extends classical stacked generalization by replacing the ordinary meta-learner with a robust estimator — such as a Huber-loss regressor, quantile regression, or a model trained on trimmed residuals — so that the ensemble's combination layer is resistant to outliers and noisy base-learner predi

machine learning

Robust Support Vector Machine

Robust SVM extends the standard support vector machine to resist the influence of outliers and mislabeled points. By replacing the hinge loss with a bounded or non-convex loss function — or by incorporating robust optimization constraints — it learns a decision boundary that is far less distorted by corrupted training

machine learning

Robust Voting Ensemble

Robust Voting Ensemble combines predictions from multiple base classifiers using noise-tolerant aggregation — such as weighted voting, trimmed voting, or median-based combination — to produce final decisions that remain reliable when individual classifiers are corrupted by noisy labels, adversarial inputs, or distribut

geoscience

Rock Mass Classification

Rock mass classification is the systematic assessment of rock quality and mechanical behavior in engineering geology, combining field observations of jointing, weathering, and strength into a numerical index. Pioneered by Bieniawski (RMR system, 1974) and Barton (Q-system, 1974), these methods enable rapid site assessm

mining engineering

Rock Mass Rating

The Rock Mass Rating (RMR) system, developed by Zbigniew Bieniawski starting in 1973, is an empirical classification that characterizes rock mass quality and estimates mining and civil engineering behavior. RMR combines five measurable geotechnical parameters into a single index ranging from 0 to 100, where higher valu

model evaluation

Root Mean Squared Error

Root Mean Squared Error is a widely used metric that measures the average magnitude of prediction errors in regression models. Originating from Carl Friedrich Gauss's work on least-squares estimation (1809), RMSE quantifies how far predictions deviate from observed values by averaging the squared differences and taking

mining engineering

Rosin-Rammler Distribution

The Rosin-Rammler Distribution, introduced by Paul Rosin and Erich Rammler in 1933, is an empirical probability distribution that describes the particle size distribution of ground or crushed materials. It characterizes fineness by two parameters: the characteristic size (d-prime) and the uniformity index (n). This dis

astronomy

Rotation Curve Analysis

Galaxy rotation curve analysis is the technique of measuring how orbital velocities change with distance from the center of a galaxy. Pioneered by Vera Rubin and W. Kent Ford Jr. in 1970, rotation curves revealed one of astronomy's great mysteries: galaxies rotate too fast to be held together by their visible stars alo

cryptography

RSA Cryptosystem

RSA is a foundational public-key cryptosystem developed by Rivest, Shamir, and Adleman in 1978. It enables secure encryption and digital signatures by using a pair of mathematically linked keys: a public key for encryption and a private key for decryption. RSA's security relies on the computational difficulty of factor

cryptography

RSA Cryptosystem Analysis

RSA (Rivest–Shamir–Adleman) is a foundational asymmetric cryptosystem introduced in 1978 that enables both encryption and digital signatures using a pair of public and private keys. It remains one of the most widely deployed cryptographic algorithms in modern security infrastructure, supporting secure communication and

acoustics

RT60 Reverberation Time

RT60 (reverberation time) is the duration required for sound energy in a room to decay by 60 decibels after the source stops. Pioneered by Wallace Clement Sabine in 1900, RT60 is the most widely used single-number descriptor of room acoustic properties. It reflects how much sound is absorbed versus reflected by room su

machine learning

Rule Induction

Rule Induction, and specifically the RIPPER (Repeated Incremental Pruning to Produce Error Reduction) algorithm, is a supervised machine learning method that learns a compact set of IF-THEN classification rules from labeled training data. Introduced by William W. Cohen in 1995, RIPPER applies a separate-and-conquer str

numerical methods

Runge-Kutta Method

The Runge-Kutta Method is a family of explicit numerical techniques for solving ordinary differential equations (ODEs) developed independently by Carl Runge in 1895 and Martin Kutta in 1901. The fourth-order variant (RK4) is one of the most widely used algorithms in computational science and engineering for time-steppi

electrical engineering

S-Parameter Analysis

S-Parameters (Scattering Parameters) characterize RF and microwave networks by their transmission and reflection of voltage waves. Introduced by Kurokawa in 1965, S-parameters are ideal for high frequencies where wave effects dominate. Unlike impedance (Z), admittance (Y), or hybrid parameters, S-parameters are directl

complex systems

Sample Entropy

Sample Entropy (SampEn) is a nonlinear measure of the complexity and regularity of a time series. Introduced by Richman and Moorman in 2000 as an improvement over Approximate Entropy (ApEn), it quantifies the likelihood that similar patterns of a given length in the series remain similar when extended by one additional

biomechanics

Scaffold Porosity Analysis

Scaffold porosity analysis characterizes the pore structure of tissue engineering scaffolds, including total porosity, pore size distribution, pore shape, and pore interconnectivity. Essential for predicting cell seeding, nutrient diffusion, and mechanical properties, this quantitative approach bridges scaffold design

computer vision

Scale-Space Theory

Scale-space theory, developed by Witkin and Lindeberg, provides a principled mathematical framework for analyzing images at multiple scales simultaneously. By treating scale as an explicit dimension and using Gaussian blurring, scale-space theory enables detection and analysis of features at appropriate scales, solving

simulation

Scenario Analysis

Scenario analysis is a structured analytical approach that systematically compares system outputs across different combinations of uncertain input values. When paired with a quantitative model, it becomes a simulation — capable of stress-testing assumptions and projecting the range of plausible outcomes. Formalised in

text mining

Scientific Text Mining

Scientific text mining is a natural-language-processing pipeline applied to academic literature. Grounded in domain-specific pretrained models such as SciBERT (Beltagy et al., 2019) and SPECTER (Cohan et al., 2020), it automatically extracts hypotheses, methodologies, findings, and scholarly contributions from full-tex

deep learning

SCINet

SCINet is a deep learning architecture for multi-step time-series forecasting introduced by Liu et al. at NeurIPS 2022. Its core idea is a recursive binary-tree structure of SCI-Blocks, each of which splits an input sequence into odd- and even-indexed sub-sequences, applies convolutional filters to model cross-subseque

deep learning

Score-Based Generative Model

A score-based generative model, introduced by Yang Song and Stefano Ermon in 2019 and generalized to the stochastic differential equation (SDE) framework in 2021, learns the gradient of the data density — the score — rather than predicting noise directly, and uses it to generate new samples. It is the mathematical gene