Some methods to sampling data points from a given distribution.
-
Updated
Jul 16, 2018 - Python
Some methods to sampling data points from a given distribution.
Code library for the DRMD framework from 'DRMD: Deep Reinforcement Learning for Malware Detection under Concept Drift'.
Adaptive Rejection Sampling for Python
Monte Carlo methods with TensorFlow
Monte is a set of Monte Carlo methods in Python. The package is written to be flexible, clear to understand and encompass variety of Monte Carlo methods.
Implementation for bayesian network
Python SDK for Quey — verifiable randomness from physical entropy. NIST-validated, Ed25519-signed draws.
Speculative decoding runtime with rejection sampling, adaptive gamma strategy, and provable correctness guarantees. Achieves 1.41x speedup on CPU with Qwen2-0.5B/1.5B pair. Draft model generates candidates, target model verifies in single forward pass. 31/31 tests passing.
Recursos sobre manejo de la incertidumbre y probabilidad por un agente inteligente, módulo de Modelos de Inteligencia Artificial
python implementation for rejection sampling and importance sampling
Agentic RL 最小实验台:把工具调用任务的成败做成可验证奖励(全部规则判定、不用模型当裁判),跑拒绝采样加 SFT 与 DPO。测试集达标率从 42.5% 提升至 62.5%
A Python implementation of Bayesian Networks from scratch, featuring exact inference (Variable Elimination) and approximate inference algorithms (Rejection Sampling, Gibbs Sampling, and Likelihood Weighting).
Application of rejection sampling and markov chain monte carlo (MCMC) algorithms to approximate bayesian computation (ABC). The project includes application of ABC to model the pharmacokinetics of theophylline.
Official PyTorch implementation of "LSRS: Latent Scale Rejection Sampling for Visual Autoregressive Modeling". An efficient test-time scaling strategy to enhance VAR image generation quality with minimal overhead.
CPU-only study of reward-model over-optimisation (Goodharting) in RLHF-style rejection-sampling self-improvement: optimise a verifiable reward and accuracy climbs; optimise a learned reward and the proxy climbs while true accuracy collapses.
无标注任务上的 Function Calling 能力后训练:程序化判卷器 + 配方驱动造题 + Agent 轨迹拒绝采样 + LoRA SFT。工具调用率 0→100%
Implementation of Prior Sampling, Rejection Sampling, Likelihood Weighting, and Gibbs Sampling for Bayesian Network Stochastic Inference.
Implementation of Prior, Rejection, Likelihood and Gibbs Sampling
Poker test for independence, Inversion method, Method of approximations, Rejection method, Quadratic congruent random number generator, Freedman–Diaconis rule, Fixation Index, Extended Haplotype Homozygosity, Wright-Fisher model
To associate your repository with the rejection-sampling topic, visit your repo's landing page and select "manage topics."