Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
133 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
46 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Adaptive Annealed Importance Sampling with Constant Rate Progress (2306.15283v1)

Published 27 Jun 2023 in stat.ML and cs.LG

Abstract: Annealed Importance Sampling (AIS) synthesizes weighted samples from an intractable distribution given its unnormalized density function. This algorithm relies on a sequence of interpolating distributions bridging the target to an initial tractable distribution such as the well-known geometric mean path of unnormalized distributions which is assumed to be suboptimal in general. In this paper, we prove that the geometric annealing corresponds to the distribution path that minimizes the KL divergence between the current particle distribution and the desired target when the feasible change in the particle distribution is constrained. Following this observation, we derive the constant rate discretization schedule for this annealing sequence, which adjusts the schedule to the difficulty of moving samples between the initial and the target distributions. We further extend our results to $f$-divergences and present the respective dynamics of annealing sequences based on which we propose the Constant Rate AIS (CR-AIS) algorithm and its efficient implementation for $\alpha$-divergences. We empirically show that CR-AIS performs well on multiple benchmark distributions while avoiding the computationally expensive tuning loop in existing Adaptive AIS.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (41)
  1. Annealed flow transport monte carlo. In International Conference on Machine Learning, pp. 318–330. PMLR, 2021.
  2. Annealed Importance Sampling with q-paths. arXiv preprint arXiv:2012.07823, 2020.
  3. Importance weighted autoencoders. In Bengio, Y. and LeCun, Y. (eds.), 4th International Conference on Learning Representations, ICLR 2016, San Juan, Puerto Rico, May 2-4, 2016, Conference Track Proceedings, 2016.
  4. An introduction to Sequential Monte Carlo. Springer, 2020.
  5. Probability functional descent: A unifying perspective on GANs, variational inference, and reinforcement learning. In International Conference on Machine Learning, pp. 1213–1222. PMLR, 2019.
  6. Dai, B. Learning over functions, distributions and dynamics via stochastic optimization. PhD thesis, Georgia Institute of Technology, 2018.
  7. Provable Bayesian inference via particle mirror descent. In Artificial Intelligence and Statistics, pp.  985–994. PMLR, 2016.
  8. Annealed Stein variational gradient descent. arXiv preprint arXiv:2101.09815, 2021.
  9. Sequential Monte Carlo samplers. Journal of the Royal Statistical Society: Series B (Statistical Methodology), 68(3):411–436, 2006.
  10. Importance weighting and variational inference. Advances in neural information processing systems, 31, 2018.
  11. Score-based diffusion meets annealed importance sampling. Advances in Neural Information Processing Systems, 35:21482–21494, 2022a.
  12. Annealed importance sampling meets score matching. In ICLR Workshop on Deep Generative Models for Highly Structured Data, 2022b.
  13. Parallel tempering: Theory, applications, and new perspectives. Physical Chemistry Chemical Physics, 7(23):3910–3916, 2005.
  14. Rethinking the effective sample size. arXiv preprint arXiv:1809.04129, 2018.
  15. Approximate inference with wasserstein gradient flows. In International Conference on Artificial Intelligence and Statistics, pp.  2581–2590. PMLR, 2020.
  16. MCMC variational inference via uncorrected hamiltonian annealing. Advances in Neural Information Processing Systems, 34:639–651, 2021.
  17. Simulating normalizing constants: From importance sampling to bridge sampling to path sampling. Statistical science, pp.  163–185, 1998.
  18. Optimization of annealed importance sampling hyperparameters. In Machine Learning and Knowledge Discovery in Databases: European Conference, ECML PKDD 2022, Grenoble, France, September 19–23, 2022, Proceedings, Part V, pp.  174–190. Springer, 2023.
  19. Annealing between distributions by averaging moments. In NIPS, pp.  2769–2777. Citeseer, 2013.
  20. Sandwiching the marginal likelihood using bidirectional Monte Carlo. CoRR, abs/1511.02543, 2015. URL http://arxiv.org/abs/1511.02543.
  21. Measuring the reliability of MCMC inference with bidirectional Monte Carlo. Advances in Neural Information Processing Systems, 29, 2016.
  22. Inference for Lévy-driven stochastic volatility models via adaptive Sequential Monte Carlo. Scandinavian Journal of Statistics, 38(1):1–22, 2011.
  23. Towards automatic model comparison: An adaptive Sequential Monte Carlo approach. 2015.
  24. Kiwaki, T. Variational optimization of annealing schedules. arXiv preprint arXiv:1502.05313, 2015.
  25. Kong, A. A note on importance sampling using standardized weights. University of Chicago, Dept. of Statistics, Tech. Rep, 348, 1992.
  26. Wasserstein proximal of GANs. In International Conference on Geometric Science of Information, pp.  524–533. Springer, 2021.
  27. Liu, Q. Stein variational gradient descent as gradient flow. Advances in neural information processing systems, 30, 2017.
  28. Stein variational gradient descent: A general purpose Bayesian inference algorithm. Advances in neural information processing systems, 29, 2016.
  29. Filtering variational objectives. Advances in Neural Information Processing Systems, 30, 2017.
  30. The thermodynamic variational objective. Advances in Neural Information Processing Systems, 32, 2019.
  31. q-paths: Generalizing the geometric annealing path using power means. In Uncertainty in Artificial Intelligence, pp.  1938–1947. PMLR, 2021.
  32. Variational Sequential Monte Carlo. In International conference on artificial intelligence and statistics, pp.  968–977. PMLR, 2018.
  33. Neal, R. M. Sampling from multimodal distributions using tempered transitions. Statistics and computing, 6(4):353–366, 1996.
  34. Neal, R. M. Annealed Importance Sampling. Statistics and computing, 11(2):125–139, 2001.
  35. Ogata, Y. A Monte Carlo method for high dimensional integration. Numerische Mathematik, 55(2):137–157, 1989.
  36. Stochastic backpropagation and approximate inference in deep generative models. In International conference on machine learning, pp. 1278–1286. PMLR, 2014.
  37. On the quantitative analysis of deep belief networks. In Proceedings of the 25th international conference on Machine learning, pp.  872–879, 2008.
  38. Monte Carlo variational auto-encoders. In International Conference on Machine Learning, pp. 10247–10257. PMLR, 2021.
  39. Stochastic normalizing flows. arXiv preprint arXiv:2002.06707, 2020.
  40. On the quantitative analysis of decoder-based generative models. In International Conference on Learning Representations, 2017.
  41. Differentiable annealed importance sampling and the perils of gradient noise. Advances in Neural Information Processing Systems, 34:19398–19410, 2021.
Citations (6)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now