按方法、领域与证据探索科学。

Magnetotellurics

Magnetotellurics (MT) is a passive geophysical method that uses natural variations in Earth's magnetic and electric fields to characterize subsurface electrical conductivity. Developed by Louis Cagniard in 1953, MT measures the impedance relationship between naturally occurring magnetic fluctuations (from solar wind an

aerospace

Mahony Filter

The Mahony Filter is a complementary observer-based attitude filter that fuses gyroscope, accelerometer, and magnetometer measurements to estimate quaternion orientation. Developed by Robert Mahony and colleagues in 2008, the filter combines gyroscope rate integration with corrective feedback from vector measurements (

ensemble learning

Majority Voting

Majority voting is an ensemble method that combines predictions from multiple base classifiers by selecting the class that receives the most votes. Each base classifier casts one vote for a predicted class, and the final prediction is the class with the majority (plurality). This approach was formalized by Leo Breiman

deep learning

Mamba (State Space Model)

Mamba is a sequence model architecture introduced by Gu and Dao in 2023 that achieves linear-time complexity while maintaining strong performance on language modeling tasks. By combining state space models with input-dependent selectivity, Mamba addresses the quadratic complexity of transformers while preserving modeli

topology

Mapper Algorithm

The Mapper algorithm is a method in topological data analysis (TDA) that produces a graph-based summary of the shape of high-dimensional point cloud data. Introduced by Singh, Mémoli, and Carlsson in 2007 at the Eurographics Symposium on Point-Based Graphics, Mapper constructs a simplicial complex — typically a graph —

biomechanics

Markerless Motion Capture

Markerless motion capture infers the 3D positions and joint angles of a moving subject from video sequences using computer vision and machine learning. Pioneered by deep learning approaches such as OpenPose and MediaPipe, it eliminates the need for reflective markers or inertial sensors, making motion capture accessibl

simulation

Markov Chain Monte Carlo

Markov Chain Monte Carlo (MCMC) is a family of simulation algorithms that constructs a Markov chain whose stationary distribution is the target posterior, enabling Bayesian inference and high-dimensional integral computation that would otherwise be analytically intractable. Pioneered by Metropolis and colleagues in 195

simulation

Markov Model

A Markov Model represents a system as a finite set of states and specifies the probability of moving from one state to another at each time step. By capturing only the current state — not the full history — it enables tractable analysis of complex dynamic processes across health economics, engineering reliability, oper

machine learning

MARS

Multivariate adaptive regression splines, introduced by Jerome Friedman in 1991, is a flexible nonparametric regression method that automatically models nonlinearities and interactions by combining piecewise-linear 'hinge' functions. It builds the model in a forward stagewise pass that adds basis functions where they h

oceanography

Marxan MPA Planning

Marxan is a decision-support system that uses optimization algorithms to design cost-effective marine protected area (MPA) networks that achieve conservation targets while minimizing socioeconomic costs. Developed by Ian Ball and Hugh Possingham in 2000, Marxan has become the global standard tool for systematic conserv

deep learning

Mask R-CNN

Mask R-CNN is a deep learning framework for instance segmentation introduced by Kaiming He, Georgia Gkioxari, Piotr Dollár, and Ross Girshick at Facebook AI Research (FAIR) in 2017. It extends Faster R-CNN by adding a parallel branch that predicts a binary pixel-level mask for each detected object instance, enabling si

deep learning

Masked Autoencoders

Masked Autoencoders (MAE) is a self-supervised learning approach introduced by He et al. in 2021 that masks random patches of an image and trains a model to reconstruct the missing content. Adapting the masked language modeling paradigm from NLP to vision, MAE learns rich visual representations by solving a challenging

machine learning

Matrix Completion

Matrix Completion is a technique for recovering a low-rank matrix from a small, possibly random subset of its entries. Introduced by Emmanuel Candès and Benjamin Recht in 2009, it reformulates the problem as nuclear norm minimization — a convex surrogate for rank minimization — and provides theoretical guarantees that

particle physics

Matrix Element Method

The Matrix Element Method (MEM) is a powerful analysis technique that leverages quantum field theory amplitudes to extract maximum physics information from individual events. By comparing observed detector signatures to predictions from matrix elements, MEM provides unbiased, model-independent measurements with excelle

meteorology

Maximum Covariance Analysis

Maximum covariance analysis (MCA) is a statistical technique that identifies coupled patterns of variability between two spatially distributed fields (e.g., sea surface temperature and precipitation). Unlike EOF analysis which focuses on variance in a single field, MCA identifies spatial patterns that are maximally cor

thermodynamics

Maximum Power Point Tracking

Maximum Power Point Tracking (MPPT) is a control algorithm for photovoltaic and wind energy systems that continuously adjusts the electrical load to extract maximum power regardless of changing irradiance and temperature. Without MPPT, a solar panel or wind turbine operates below its power potential due to impedance mi

mining engineering

McCabe-Thiele Method

The McCabe-Thiele Method, introduced by Warren L. McCabe and Ernest W. Thiele in 1925, is a graphical technique for designing and analyzing distillation columns. It predicts the number of theoretical plates (stages) needed to achieve a desired separation between light and heavy components. While primarily a chemical en

genetics

McDonald-Kreitman Test

The McDonald-Kreitman (MK) test is a statistical method for detecting adaptive evolution by comparing ratios of synonymous and nonsynonymous substitutions within and between species. Developed by James McDonald and Martin Kreitman in 1991, this test exploits the key insight that neutral mutations accumulate at similar

model evaluation

Mean Absolute Error

Mean Absolute Error is a robust metric that measures the average absolute magnitude of prediction errors in regression models. Dating back to Pierre-Simon Laplace's work on observational errors (1799), MAE quantifies typical prediction deviation by averaging the absolute differences between observed and predicted value

model evaluation

Mean Absolute Percentage Error

Mean Absolute Percentage Error measures prediction accuracy as a percentage relative to actual values, expressing errors in units that are scale-independent and interpretable across datasets. Formalized by J. Scott Armstrong in 1985, MAPE is widely used in forecasting, supply chain, and business analytics where results

model evaluation

Mean Absolute Scaled Error

Mean Absolute Scaled Error is a scale-independent metric that measures prediction accuracy relative to a simple baseline (naive forecast). Introduced by Hyndman and Koehler (2006), MASE directly compares model performance to a reference method, overcoming limitations of MAPE and other percentage-based metrics.

machine learning

Mean Shift

Mean Shift is a non-parametric, iterative mode-seeking algorithm that identifies clusters as the peaks of an underlying probability density function. Originally introduced by Fukunaga and Hostetler (1975) for gradient estimation in pattern recognition, it was substantially extended and popularized by Comaniciu and Meer

model evaluation

Mean Squared Error

Mean Squared Error is the foundational loss function for regression models, measuring the average squared deviation between predictions and observations. Originating from Gauss and Legendre's method of least squares (1805-1809), MSE is the basis for ordinary least squares regression and remains central to modern machin

music information retrieval

Melody Extraction

Melody extraction is the task of automatically isolating the main melodic contour from polyphonic music recordings. It originated from music transcription research in the 2000s and addresses the core challenge of human pitch perception: identifying the perceptually dominant pitch when many instruments play simultaneous

bioinformatics

Metabolomics analysis

Metabolomics analysis is the large-scale, systematic measurement of small-molecule metabolites in a biological sample to characterise the metabolome — the complete set of metabolic intermediates and products present under defined conditions. By coupling high-throughput analytical platforms such as mass spectrometry (MS

bioinformatics

Metagenomic Binning

Metagenomic binning partitions assembled contigs from complex microbial communities into distinct genome bins, each representing an individual organism or strain. Pioneered by Banfield and colleagues, this pipeline isolates single-organism genomes (metagenome-assembled genomes or MAGs) from environmental samples withou

electrical engineering

Method of Moments

The Method of Moments (MoM) is a powerful numerical technique for solving electromagnetic boundary integral equations derived from Maxwell equations. Pioneered by Roger Harrington in 1968, MoM discretizes only radiating surfaces and boundaries (antennas, conductors, dielectrics), not the surrounding space, making it ef

machine learning

Metric Learning

Metric learning is a machine-learning framework that trains a distance or similarity function from data so that semantically similar examples end up close together in the learned space while dissimilar examples are pushed apart. Unlike fixed distances such as Euclidean, the learned metric adapts to the structure of the

applied physics

MFCC

Mel-Frequency Cepstral Coefficients (MFCCs) are a compact representation of audio features that mimic human auditory perception. Introduced by Davis and Mermelstein in 1980, MFCCs are the de facto feature extraction method for speech recognition and environmental sound analysis. They compress the frequency information

deep learning

MICN

MICN (Multi-scale Isometric Convolution Network) is a convolutional neural network architecture for long-term time-series forecasting introduced by Huiqiang Wang and colleagues at ICLR 2023. Its central idea is to capture both local temporal patterns and global seasonal dependencies simultaneously through multi-scale i

model evaluation

Micro-averaged F1

Micro-averaged F1 computes the F1-score by aggregating true positives, false positives, and false negatives across all classes, then calculating a single metric. It is equivalent to accuracy in multi-class classification and is useful when class distributions reflect their natural importance.

simulation

Microsimulation

Microsimulation is a computational method that simulates policy effects by operating directly on a population of individual micro-units — households, firms, patients — and applying rules to each unit according to its own demographic, economic, and behavioural characteristics. Developed conceptually by Guy Orcutt in 195

telecommunications

MIMO

MIMO is a technique that uses multiple transmit and receive antennas to significantly increase channel capacity and reliability. Pioneered theoretically by Telatar (1999) and Foschini & Gans (1998), MIMO exploits multipath propagation—typically a liability in wireless—as an asset by creating independent spatial channel

mining engineering

Mine Ventilation

Mine ventilation is the design and operation of systems that deliver fresh air to underground mining areas and remove contaminated air, heat, and hazardous gases. It is critical for worker safety and productivity, maintaining breathable air (sufficient oxygen, low dust and gas concentrations) and acceptable temperature

linguistics

Minimalist Program

The Minimalist Program (MP) is a framework for generative syntax developed by Noam Chomsky in 1995, designed to explain linguistic structure while assuming the fewest possible theoretical mechanisms. The program seeks principles that are simple, elegant, and motivated by language evolution. It addresses core questions:

particle physics

Missing Transverse Energy

Missing transverse energy (MET) is a powerful technique used in high-energy physics to infer the presence of invisible particles, primarily neutrinos, that escape a detector without leaving a trace. By measuring the imbalance of transverse momentum in the event, physicists can detect signatures of weakly interacting pa

deep learning

Mixture of Experts

Mixture of Experts (MoE) is a sparse neural-network architecture, introduced by Shazeer and colleagues in 2017 with the sparsely-gated MoE layer, in which only a subset of expert sub-networks is activated for each input. As seen in models such as Switch Transformer and Mixtral, it holds computation cost fixed even as t

deep learning

MobileNet

MobileNet is a family of lightweight convolutional neural network architectures introduced by Howard et al. at Google in 2017. It is designed to run image classification, object detection, and other vision tasks directly on mobile devices and embedded systems with limited computational budgets. By replacing standard co

manufacturing

Modal Analysis

Modal analysis is a computational and experimental method for determining the natural frequencies and associated mode shapes of a mechanical structure. By decomposing structural vibration into its fundamental modes (natural oscillation patterns), engineers can predict resonance frequencies, assess dynamic response to e

machine learning

Model Calibration

Model calibration is a post-hoc technique that adjusts the probability outputs of a trained classifier so that predicted confidence scores match empirical outcome frequencies. A classifier is said to be perfectly calibrated if, among all predictions made with confidence p, exactly a fraction p of them are correct. Syst

control theory

Model Predictive Control

Model Predictive Control (MPC) is an advanced control strategy that uses an explicit process model to predict future system behavior over a finite horizon and solves an optimization problem at each control step. First formalized by Richalet et al. in 1978, MPC has become the dominant approach in process control industr

civil engineering

MODFLOW Groundwater Modeling

MODFLOW is the U.S. Geological Survey's open-source, modular finite-difference model for simulating three-dimensional groundwater flow through porous media. First released in 1984 and continuously updated — most recently as MODFLOW-6 — it is the global standard for quantitative hydrogeological analysis, widely used in

network analysis

Modularity Analysis

Modularity analysis is a network science method, formalized by Newman and Girvan in 2004, that detects community structure in graphs by measuring whether edges are more concentrated within groups than expected by chance. Its scalar quality index Q guides algorithms that partition nodes into cohesive clusters, making it

time series

MODWT

The maximal overlap discrete wavelet transform (MODWT) is a translation-invariant wavelet decomposition method that addresses a key limitation of the standard DWT: lack of shift invariance. Introduced by Percival and Walden (1995), MODWT applies the same wavelet filters at each scale without downsampling, producing an

deep learning

Moirai

Moirai is a foundation model for universal time-series forecasting introduced by Gerald Woo and colleagues at Salesforce Research in 2024 and presented at ICML. The core idea is to pre-train a single large Transformer on an exceptionally diverse corpus of time-series data (LOTSA) spanning many domains and frequencies,

bioinformatics

Molecular Docking

Molecular docking predicts the preferred binding orientation and affinity of a ligand (small molecule) within a protein binding pocket. Pioneered by Kuntz and colleagues in 1982, this computational method searches conformational space to find energetically favorable ligand-protein complexes, enabling rapid screening of

materials science

Molecular Dynamics

Molecular Dynamics (MD) is a computational technique that simulates the motion of atoms and molecules by solving Newton's equations of motion under specified forces. Pioneered by Alder and Wainwright in 1957, MD integrates time-dependent atomic trajectories from initial positions, allowing prediction of material proper

quantum computing

Moller-Plesset Perturbation Theory

Möller-Plesset perturbation theory is a post-Hartree-Fock method that systematically corrects the HF reference by treating electron correlation as a perturbation. Introduced in 1934, MP theory provides increasingly accurate energy estimates (MP2, MP3, MP4, ...) by expanding the correlation energy in orders of perturbat

meteorology

Monin-Obukhov Similarity

Monin-Obukhov similarity theory is a fundamental framework in boundary layer meteorology that describes how wind speed, temperature, and humidity vary with height near the surface. Published in 1954, it shows that normalized vertical profiles depend on a single dimensionless parameter—the Monin-Obukhov stability parame

nuclear physics

Monte Carlo Neutron & Particle Transport

Monte Carlo neutron and particle transport is a stochastic simulation method that tracks individual particle histories through matter, developed by Metropolis and Ulam in 1949 during the Manhattan Project. By sampling random numbers to determine collision locations, energy transfers, and scattering angles, it produces

electrical engineering

Monte Carlo Process Variation

Monte Carlo Process Variation analysis quantifies the impact of manufacturing uncertainties on circuit performance using statistical sampling. As semiconductor technology scales, process variations (gate length, oxide thickness, dopant fluctuations) create significant uncertainties in delay, power, and leakage. Monte C

text mining

Morphological Analysis

Morphological analysis splits words into their stems and affixes so that different surface forms of the same word can be treated as one. It covers two complementary approaches — rule-based stemming, such as the Porter (1980) and Snowball algorithms, and dictionary-aware lemmatization — and is a critical text-normalisat

electrical engineering

Motor Drive Efficiency Analysis

Motor drive efficiency analysis quantifies energy losses in electrical motors and variable frequency drive (VFD) systems, which together comprise the largest industrial electrical load. Methods assess copper losses in windings, core (iron) losses, mechanical losses, and converter losses to identify efficiency improveme

telecommunications

MPLS

Multiprotocol Label Switching (MPLS) is a forwarding paradigm that prepends a short label to packets, enabling routers to make forwarding decisions based on the label rather than IP destination address. Introduced by IETF (2001), MPLS was designed to enable traffic engineering, VPN creation, and fast rerouting in IP ne

text mining

Multi-Document Summarization

Multi-document summarization (MDS) is a natural-language-processing task that condenses a cluster of related documents into a single comprehensive, coherent, and non-redundant summary. Formally described by Erkan and Radev (2004) through the LexRank algorithm, MDS is used in news cluster analysis, systematic literature

machine learning

Multi-layer Perceptron

The Multi-layer Perceptron (MLP) is a feedforward neural network architecture trained by backpropagation, formalised by Rumelhart, Hinton, and Williams in their landmark 1986 Nature paper. Composed of an input layer, one or more hidden layers of neurons with nonlinear activation functions, and an output layer, the MLP

simulation

Multi-objective agent-based modeling

Multi-Objective Agent-Based Modeling (MO-ABM) couples agent-based simulation with multi-objective optimization to simultaneously optimize several conflicting performance criteria across complex adaptive systems. Autonomous agents interact according to behavioral rules while an optimizer searches for parameter configura

simulation

Multi-objective cellular automata

Multi-Objective Cellular Automata (MOCA) couples the bottom-up spatial dynamics of cellular automata with multi-objective optimization to simultaneously pursue competing goals — such as maximizing urban compactness while minimizing ecosystem loss. Each grid cell updates its state based on transition rules that are cali

simulation

Multi-objective discrete-event simulation

Multi-Objective Discrete-Event Simulation (MO-DES) couples a discrete-event simulation engine with multi-objective optimization to explore trade-offs among two or more conflicting performance measures — such as throughput, cost, and waiting time — across stochastic, time-ordered process models. It is widely applied in

simulation

Multi-objective Markov Model

A Multi-objective Markov Model (MOMDP) extends classical Markov Decision Processes to settings where an agent must optimize several reward signals simultaneously. Instead of a single optimal policy, the model produces a Pareto-optimal set of policies, enabling decision-makers to navigate trade-offs between competing go