This paper reviews recent advancements in amortized Variational Inference.
problem Scalability and efficiency issues in traditional Variational Inference.
method Systematic review of various Variational Inference techniques, focusing on amortized approaches.
result Amortized Variational Inference improves scalability and efficiency for generative modeling tasks.
This work proposes a new method for variational inference using Wasserstein gradient descent.
problem Optimizing variational parameters to match a true posterior distribution.
method Reinterpreting VI as an optimization problem over a variational parameter space, using Wasserstein gradient descent.
result The proposed Wasserstein gradient descent can be seen as a generalization of existing optimization techniques in VI.
Paper introduces f-divergence variational inference for broader application.
problem Variational inference limited to specific divergences.
method Generalizes variational inference to all f-divergences using f-divergence minimization.
result Unified framework for variational inference with arbitrary f-divergences.
Novel approach simplifies VI problems with faster performance.
problem Black-box VI optimization problems.
method Sample Average Approximation (SAA) combined with quasi-Newton methods and line search.
result Achieves faster performance than existing methods.
BF-VI improves posterior approximation in complex models.
problem Inefficient posterior approximations in complex models.
method Combines normalizing flows and Bernstein polynomial transformations.
result BF-VI outperforms other VI methods in approximating complex multivariate posteriors.
VI approximates complex densities faster than classical methods.
problem Approximating complex probability densities.
method Optimization of a family of probability density functions using KL divergence.
result VI converges faster than Markov Chain Monte Carlo.
New algorithms solve stochastic variational inequalities without bounded variance assumption.
problem Solving stochastic variational inequalities without bounded variance assumption.
method Developed algorithms for two classes of problems: monotone and structured nonmonotone VIs.
result Oracle complexity of O(ε^-4) for solving VIs with unbounded domains and possibly unbounded variance.
Paper advances black-box VI using flows and Monte-Carlo methods.
problem Improving automatic posterior inference in black-box VI.
method Combines normalizing flows, Monte-Carlo methods, and optimization considerations.
result Significant improvement in state-of-the-art variational inference.
A new variational inference method using Gaussian score matching.
problem Approximating posterior distributions in Bayesian statistics.
method Score matching principle applied to variational inference.
result Gaussian score matching VI (GSM-VI) is faster and requires fewer gradient evaluations.
A new differentiable model for sampling DAGs that speeds up optimization.
problem Efficiently sampling and learning DAG structures in continuous optimization.
method DP-DAG model with VI-DP-DAG for DAG learning from data.
result VI-DP-DAG outperforms other methods in DAG structure and causal mechanism learning.
A-VI can approximate F-VI under certain conditions, improving inference in some models.
problem Approximate Bayesian inference using factorized distributions.
method Amortized variational inference (A-VI) learns a common inference function for latent variables.
result A-VI can close the amortization gap in simple hierarchical models.
New algorithm radVI improves variational inference by optimizing radial profiles.
problem Gaussian approximations often fail to capture the radial profile of complex distributions.
method Optimizes over radial profiles in variational inference, providing theoretical guarantees.
result Theoretical convergence guarantees for radVI, improving over existing VI methods.
One of the core problems of modern statistics is to approximate difficult-to-compute probability densities. This problem is especially important in Bayesian statistics, which frames all inference about unknown quantities as a calculation involving the posterior density. In this paper, we review variational inference (V…
Flexible Bayesian approach for generalized linear models, especially for sparse logistic regression.
problem Sparse logistic regression challenges in machine learning.
method Empirical Bayes approach with mean-field variational inference, tuning-free and scalable.
result Superior predictive performance in sparse logistic regression compared to existing methods.
A new VIS approach improves log-likelihood estimation in latent variable models.
problem Challenges in achieving high log-likelihood with VI for complex posterior distributions.
method Uses forward χ 2 χ^2 χ 2 divergence to optimize proposal distribution for better log-likelihood estimation. result Consistently outperforms state-of-the-art baselines in log-likelihood and parameter estimation.
Many modern unsupervised or semi-supervised machine learning algorithms rely on Bayesian probabilistic models. These models are usually intractable and thus require approximate inference. Variational inference (VI) lets us approximate a high-dimensional Bayesian posterior with a simpler variational distribution by solv…
Combines VI and EP for better Gaussian process hyperparameter learning.
problem Improving hyperparameter learning in Gaussian processes for better performance.
method Hybrid training procedure combining Variational Inference (VI) for posterior inference and Expectation Propagation (EP) for hyperparameter learning.
result The hybrid training procedure provides a better learning objective and generalizes better than using only VI or EP.
A new method interpolates between sampling and variational inference using stochastic mixtures.
problem Combining the strengths of sampling and variational inference methods.
method Develops a framework using stochastic mixtures of simple component distributions to interpolate between sampling and variational inference.
result Improves on both sampling and variational inference methods by reducing bias and variance.
CAVI converges globally or locally exponentially for two-block models.
problem Convergence analysis of coordinate ascent variational inference (CAVI).
method Analysis of CAVI convergence using functional analysis and optimization.
result General conditions for certifying global or local exponential convergence of CAVI.
New methods improve multi-agent reinforcement learning by addressing rotational dynamics.
problem Reproducibility crisis in multi-agent reinforcement learning.
method Reframing MARL approaches using Variational Inequalities (VIs) and proposing gradient-based VI methods.
result Significant performance improvements across benchmarks, including better convergence to equilibrium strategies in zero-sum games.
New method improves approximate inference for Bayesian models.
problem Approximate inference for high-dimensional Bayesian models.
method Entropic regularization of mean-field variational inference.
result Improved recovery of true posterior dependency.
Develops possibilistic VI using maxitive Donsker-Varadhan formulation.
problem Adapting VI to possibilistic framework to handle uncertainty.
method Maxitive Donsker-Varadhan formulation for possibilistic VI.
result CBOpt optimizers achieve competitive performance in image classification.
Simulation-free VI closes the approximation gap in latent SDEs
problem Recovering dynamical systems from noisy observations
method Helmholtz-SDE
result Recovers dynamics more faithfully than prior methods
This paper compares gradient estimators in importance-weighted VI and justifies the superiority of DREP over REP.
problem Understanding the impact of gradient estimators on importance-weighted VI algorithms.
method Unified theoretical comparison of reparameterized and doubly-reparameterized gradient estimators tied to IWAE, VR, and VR-IWAE bounds.
result Formally justifies the superiority of doubly-reparameterized gradient estimators over reparameterized ones in importance-weighted VI.
This study analyzes theoretical guarantees for VI with fixed-variance Gaussian mixtures.
problem Theoretical analysis of variational inference with non-Gaussian distributions.
method Formulates variational inference as minimizing a mollified relative entropy, solving it through gradient descent on particle positions.
result Establishes descent lemma and approximation error bounds for optimization of variational inference.
Variational inference (VI) provides fast approximations of a Bayesian posterior in part because it formulates posterior approximation as an optimization problem: to find the closest distribution to the exact posterior over some family of distributions. For practical reasons, the family of distributions in VI is usually…
New algorithm speeds up large-scale statistical inference.
problem Efficiently solving large-scale mean-field variational inference problems.
method Developed a novel primal-dual algorithm (PD-VI) and a block-preconditioned extension (P 2 ^2 2 D-VI) for mean-field variational inference. result PD-VI and P 2 ^2 2 D-VI achieve faster convergence and better solution quality compared to existing methods. YOASOVI improves stochastic VI for large models with fast, self-correcting sampling.
problem Efficiently performing stochastic Variational Inference on large Bayesian models.
method YOASOVI uses acceptance sampling to draw only one sample per iteration, improving convergence speed and accuracy.
result YOASOVI converges faster and more accurately than regular Monte Carlo and Quasi-Monte Carlo methods.
Recent work in variational inference (VI) uses ideas from Monte Carlo estimation to tighten the lower bounds on the log-likelihood that are used as objectives. However, there is no systematic understanding of how optimizing different objectives relates to approximating the posterior distribution. Developing such a conn…
DAIS minimizes symmetrized KL divergence between initial and target distributions.
problem Optimizing over initial distributions in importance sampling.
method Differentiable annealed importance sampling (DAIS) minimizing symmetrized KL divergence.
result DAIS minimizes symmetrized KL divergence between initial and target distributions.
This paper shows how to learn variational inequalities fast with strong monotonicity.
problem Learning variational inequalities efficiently.
method Extending convex optimization techniques to variational inequalities with strong monotonicity.
result Fast generalization rates of Θ ( 1 / ε ) Θ(1/ε) Θ ( 1/ ε ) for learning variational inequalities. Uncertainty computation in deep learning is essential to design robust and reliable systems. Variational inference (VI) is a promising approach for such computation, but requires more effort to implement and execute compared to maximum-likelihood methods. In this paper, we propose new natural-gradient algorithms to red…
Equivalence of convex optimization, saddle-point problems, and variational inequalities is a well-established concept. The variational inequality (VI) is a static problem which is studied under dynamical settings using a framework called the projected dynamical system, whose stationary points coincide with the static s…
CAVI converges for log-concave measures via optimal transport.
problem Finding the closest product measure to a log-concave measure via CAVI.
method Adapting coordinate descent techniques from Euclidean space to optimal transport for log-concave densities.
result Proves convergence of CAVI for log-concave densities and provides rates of convergence under additional conditions.
RVI accelerates encoderless VI for faster convergence.
problem Slow convergence in encoderless VI methods.
method Introduces Relay Variational Inference (RVI) for faster learning.
result RVI outperforms existing methods in convergence speed and performance.
Improves hyperparameter learning in GP models with non-conjugate likelihoods.
problem Hyperparameter learning entangled with approximate inference in GP models.
method Hybrid training procedure combining VI for inference and EP-like marginal likelihood approximation for hyperparameter learning.
result Empirically demonstrates the effectiveness of the proposed training procedure across various data sets.
Automates VI divergence selection for efficient few-shot learning.
problem Efficiently selecting divergence measures for VI to improve performance.
method Meta-learning algorithm to learn optimal divergence metric and variational parameter initialization.
result Meta-learning approach outperforms standard VI methods across various tasks.
We develop a parallel variational inference (VI) procedure for use in data-distributed settings, where each machine only has access to a subset of data and runs VI independently, without communicating with other machines. This type of "embarrassingly parallel" procedure has recently been developed for MCMC inference al…
Two-stage nonconvex algorithm and convex relaxation both achieve optimal accuracy in noisy blind deconvolution.
problem Solving bilinear systems of equations with random noise under different designs.
method Two-stage nonconvex algorithm and convex relaxation.
result Both methods achieve minimax-optimal accuracy in the presence of random noise.
Explaining how overparametrized neural networks simultaneously achieve low risk and zero empirical risk on benchmark datasets is an open problem. PAC-Bayes bounds optimized using variational inference (VI) have been recently proposed as a promising direction in obtaining non-vacuous bounds. We show empirically that thi…
VPR improves posterior uncertainty quantification by combining VI and predictive resampling.
problem Inaccurate posterior sampling with MCMC due to computational constraints.
method Variational predictive resampling (VPR) that uses VI's predictive strength and imputes future observations.
result VPR converges to the exact Bayesian posterior in a Gaussian location model and improves uncertainty quantification.
New findings on Malgrange-Galois groupoid for Painlevé VI equation parameters.
problem Understanding transformations preserving specific forms for Painlevé VI equation.
method Computed Malgrange-Galois groupoid for Painlevé VI family with all parameters.
result Solutions of Painlevé VI do not satisfy new partial differential equations.
Analyzes factors affecting flow VI performance.
problem Consistent performance of flow VI across studies.
method Step-by-step analysis of capacity, objectives, batchsize, estimators, and step-sizes.
result Specific recommendations and a flow VI recipe.
Improved VI with Price's gradient estimator for target log-density.
problem Approximating target distributions from unnormalized log-densities.
method Stochastic gradient-based variational inference with Price's gradient estimator.
result Identifies Price's gradient as the key to WVI's superior performance.
Variational inference (VI) has become the method of choice for fitting many modern probabilistic models. However, practitioners are faced with a fragmented literature that offers a bewildering array of algorithmic options. First, the variational family. Second, the granularity of the updates e.g. whether the updates ar…
Paper improves REINFORCE for VI without restrictive assumptions.
problem Improves REINFORCE for VI without restrictive assumptions.
method Introduces VIMCO- ⋆ \star ⋆ gradient estimator to overcome SNR collapse. result VIMCO- ⋆ \star ⋆ achieves N \sqrt{N} N SNR scaling, superior to existing VIMCO. Faster algorithms for solving multichain MDPs under average-reward criterion.
problem Navigating towards the best connected component in multichain MDPs.
method Developed algorithms to better solve the navigational subproblem, achieving faster convergence rates.
result Improved rates of convergence and sharper complexity measures for multichain MDPs.
Approximating a probability density in a tractable manner is a central task in Bayesian statistics. Variational Inference (VI) is a popular technique that achieves tractability by choosing a relatively simple variational family. Borrowing ideas from the classic boosting framework, recent approaches attempt to \emph{boo…