Pith. sign in

REVIEW 2 major objections 4 minor 58 references

Icebreaker: Element-wise Active Information Acquisition with Bayesian Deep Latent Gaussian Model

T0 review · 2 major / 4 minor · reviewed 2026-08-14 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Icebreaker, a fully Bayesian deep latent Gaussian model with partial amortized inference and uncertainty-driven acquisition, claims to solve the ice-start problem and reach the best test-time performance using the least training data.

desk verdict A solid combination paper: ice-start is a real problem, the PA-BELGAM inference scheme is a sensible engineering contribution, and the acquisition functions are well-motivated, but the theoretical grounding in Eq. 3 has a real gap that needs patching before the uncertainty estimates can be trusted. read the letter →

arxiv 1908.04537 v2 pith:3SGF5QEZ submitted 2019-08-13 cs.LG cs.AIstat.ML

Wenbo Gong , Sebastian Tschiatschek , Richard Turner , Sebastian Nowozin , José Miguel Hernández-Lobato , Cheng Zhang This is my paper · ORCID
classification cs.LGcs.AIstat.ML
keywords ice-startproblemactivefeatureacquisitionBayesiandeeplatentGaussianmodelpartialamortizedinferencestochasticgradientHamiltonianMonteCarlomissingdataimputationpredictionepistemicuncertainty
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper introduces the ice-start problem: deploying machine learning when almost no training data exists and every individual feature value costs something to obtain, as in medical tests or recommender-system ratings. It proposes Icebreaker, which combines a Bayesian deep latent Gaussian model with an inference scheme that keeps uncertainty about the model's own parameters alive, and uses that uncertainty to decide which feature elements to purchase during training. The authors report that this uncertainty-aware element-wise acquisition produces the best test-time performance with the smallest training set, outperforming previous variational-autoencoder-based methods and whole-row active learning. The reason to care is that in expensive-data settings the question is not just what to label but which measurements are worth taking at all.

What carries the argument

The machinery is partial amortized inference: local latent variables $z_i$ are encoded by a set-encoding network $q_\phi(z_i|x_i)$ that accepts variable-size partial observations, while global weights $\theta$ are produced by preconditioned stochastic gradient Hamiltonian Monte Carlo. The step that carries the argument is Eq.~3, which replaces the intractable gradient $\nabla_\theta \log p(X_O,\theta)$ with the gradient of the ELBO maximized over the amortized family $\mathcal{F}$; this makes the weight sampler feasible and supplies the model-parameter uncertainty that both acquisition functions consume. The imputation acquisition uses nested Monte Carlo estimates of expected posterior-entropy reduction, and the prediction acquisition uses the conditional variant, blended by the parameter $\alpha$.

What would settle it

Run PA-BELGAM on a small dataset such as Boston Housing and compare its stochastic gradient Hamiltonian Monte Carlo samples of the weights against a reference posterior obtained by full Hamiltonian Monte Carlo over both latent variables and weights. If the reference and approximate posterior predict the held-out data differently, or if the top-ranked candidate acquisitions from Eq.~7 disagree with the ranking computed from the reference posterior, the central claim fails.

Watch

Extended reading notes

Core claim

The paper's central discovery is that the ice-start problem can be addressed by making the training data itself the object of active acquisition. In BELGAM, each observed feature $x_{i,d}$ is generated from a local latent variable $z_i$ through a Bayesian neural network with global weights $\theta$; PA-BELGAM then infers $z_i$ with a permutation-invariant amortized encoder and samples $\theta$ with stochastic gradient Hamiltonian Monte Carlo. This gives a scalable posterior over model parameters, and the paper defines two acquisition rules: Eq.~5 scores a candidate feature by the expected reduction in posterior uncertainty of $\theta$, while Eq.~8 scores it by conditional mutual information between the target and $\theta$, with a combined objective that equals half the conditional mutual information $I(\theta; y_i,x_{i,d} \mid X_O)$ when $\alpha=1/2$. On UCI benchmarks, MovieLens-1M, and MIMIC-III, Icebreaker achieves lower NLL and AUIC than random acquisition, the partial-VAE baseline, and row-wise acquisition, and its long-tailed selection pattern shows it exploits feature correlations inside rows while still exploring new ones.

Load-bearing premise

The method assumes that replacing the true intractable parameter-gradient with the gradient of a maximized variational lower bound over the encoder family is accurate enough that the sampling step still explores nearly the true posterior over the weights; if the encoder family is too small, the uncertainty scores that decide which feature values to acquire are biased.

Editorial extensions

If this is right

  • Where training data are scarce, using a Bayesian treatment of the decoder improves imputation and active prediction even without active acquisition, so parameter uncertainty is a prerequisite for efficient cold-start learning.
  • Element-wise training acquisition is more data-efficient than acquiring whole rows: with the same query budget, Icebreaker obtains many partially observed points and reaches good test-time performance earlier than row-wise active learning.
  • The acquisition functions balance exploration and exploitation; at $\alpha=1/2$ the combined objective is exactly half the conditional mutual information between the weights and the feature-target pair, giving it a clear information-theoretic meaning.
  • Models trained on Icebreaker-selected data also make test-time active prediction more effective, lowering the area under the information curve compared with random or row-wise training acquisition.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The same partial-amortization recipe—amortize local variables, sample global weights—can likely be dropped into other deep generative models, so the contribution may help any latent-variable setting where decisions depend on parameter uncertainty.
  • A deployment of Icebreaker would benefit from a posterior-calibration check against a full Bayesian baseline before trusting the acquisition rankings, since the paper does not directly measure the gap between the approximate and true posterior over the weights.
  • The information-theoretic objectives suggest a testable extension: replacing raw mutual information with task-specific utility, such as expected health outcome, should change the acquisition pattern, and the paper leaves that comparison open.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

2 major / 4 minor

Summary. This paper introduces the 'ice-start' problem: deploying a prediction or imputation model when initially little or no training data is available and feature elements can be acquired at cost. The authors propose Icebreaker, a framework built on a Bayesian Deep Latent Gaussian Model (BELGAM) with partial amortized inference (PA-BELGAM), in which local latent variables are handled by an amortized encoder and global weights are sampled with SGHMC. Two training-time acquisition functions are proposed, one for imputation and one for active prediction, and the framework is evaluated on UCI benchmarks, MovieLens, and MIMIC-III. The paper reports that Icebreaker and PA-BELGAM outperform P-VAE and random acquisition, especially in the small-data regime.

Significance. If the result holds, this is a potentially significant contribution because it addresses element-wise training-time acquisition, a practically important problem, and it combines amortized inference with MCMC in a Bayesian deep generative model. The acquisition functions are information-theoretically motivated, and the experimental evaluation spans multiple benchmarks and real-world applications. The paper also gives a clear problem formulation and a detailed discussion of feature-selection patterns. However, the theoretical grounding depends on an unvalidated gradient identity that biases the SGHMC posterior; without a fix, the uncertainty estimates that drive the acquisition functions are not justified. The empirical comparisons also lack statistical support. The central idea is interesting, but the current manuscript needs substantial corrective work.

major comments (2)
  1. [Section 2.2, Eq. (3); Algorithm 4] The equality in Eq. (3), ∇θ log p(XO,θ) = ∇θ max_{qφ∈F} Ljoint(XO;φ), is not valid for the implemented encoder. For the right-hand side to equal the left, the maximizing qφ would have to coincide with the exact conditional posterior p(Z|XO,θ) for every relevant θ. But the inference network qφ(zi|xi) does not take θ as an input, and Eq. (4) optimizes it against an average over SGHMC samples rather than against the θ-conditional posterior. Hence the maximizer in Eq. (3) is not p(Z|XO,θ), and the SGHMC updates in Algorithm 4 use a biased gradient: E_{qφ}[∇θ log p(xi|zi,θ)] instead of E_{p(zi|xi,θ)}[∇θ log p(xi|zi,θ)]. No bound on the resulting bias is provided. Since the θ samples are used both for prediction and for the acquisition functions (Eqs. 5, 7, 8, 9, 10), the epistemic uncertainty driving Icebreaker is not justified. This is a load-bearing defect in the derivation, not merely a cosmetic issue. The authors should either use a θ-conditioned amortized family qφ(zi|xi,θ), provide an unbiased estimator of ∇θ log p(xi|θ), or explicitly treat the sampler as targeting an approximate posterior and demonstrate that the approximation is accurate enough for the proposed acquisition functions.
  2. [Section 5, Figures 4, 6, 9, 10] The abstract and Section 5 claim that BELGAM/Icebreaker performs 'significantly better' than baselines, but the main experimental curves are reported without error bars, confidence intervals, or statistical significance tests. The text states that all experiments are averaged over 10 runs, but no measure of spread is provided for the UCI or MovieLens results; only the MIMIC experiment mentions standard error. Without this information, the central empirical claim of significant improvement is not statistically supported. The authors should report standard deviations or bootstrap intervals and, where feasible, paired significance tests across runs.
minor comments (4)
  1. [Section 4.2, Eq. (8)] The text states that Eq. (8) is 'equivalent to conditional mutual information I(yi,θ|xi,d; XO)'. Expanding the expression gives H(yi|xi,d,XO) - E_θ[H(yi|θ,xi,d,XO)] = I(yi; θ | xi,d, XO), i.e., the mutual information between yi and θ conditional on xi,d and XO. The notation I(yi,θ|xi,d; XO) is non-standard and should be corrected to avoid conflating this with a three-way conditional mutual information. The Appendix C claim that the α=1/2 combination equals 1/2 I(θ; {yi,xi,d}|XO) appears consistent.
  2. [Sections 2.1 and 6] There are typos in key terms: 'epidemic uncertainty' should be 'epistemic uncertainty' (Section 2.1), 'aleoteric uncertainty' should be 'aleatoric uncertainty' (Section 6), and 'recommander' should be 'recommender' (Sections 1 and 3.1).
  3. [Section 5, experimental setup] The problem definition states Dtrain=∅ for the ice-start scenario, but the experiments initialize with 2% (UCI) or 0.5% (MovieLens/MIMIC) randomly selected pre-train data. This discrepancy should be acknowledged and discussed, since the acquisition procedure is not evaluated from a completely empty training set.
  4. [Appendix D.1, Eq. (28)] The normalization formula for the softmax weights is written ambiguously as wid = exp(rid/T) / Σ rid exp(rid/T). It should be wid = exp(rid/T) / Σ_{jd} exp(rjd/T) over the candidate elements in the pool.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: Icebreaker's acquisition gains are tested on held-out data, and its active-learning objectives are not fitted to those test targets.

full rationale

The paper's central claim—that Icebreaker achieves the best test-time performance with the fewest training feature elements—is not circular. The acquisition functions (Eqs. 5, 7, 8, 9, 10) are information-theoretic objectives computed from the current model's posterior, and the reported NLL and AUIC values are evaluated on test sets that are not used to fit any parameter of those acquisition functions. The comparison against random acquisition under the same PA-BELGAM/P-VAE backbone isolates the effect of the acquisition rule, and the test metrics are external to the selection objective. The paper does build on prior same-group work (P-VAE and EDDI, Ma et al. 2018, 2019), but these are used as architectural baselines and test-time components, not as the source of the claimed improvement; the improvement over P-VAE and random selection is demonstrated empirically on UCI, MovieLens, and MIMIC. The only notable theoretical concern is the unproven gradient identity in Eq. 3, which requires the amortized encoder to be able to represent the θ-conditional posterior; the implemented encoder qφ(zi|xi) is not θ-conditioned, so the SGHMC uncertainty estimates may be biased. This is a correctness or rigor gap, not a circularity: the paper does not define its predictions in terms of its fitted inputs, and the empirical evaluation would still be meaningful even if the posterior approximation is imperfect. Therefore the circularity score is 0.

Assumptions & free parameters 3 free parameters · 3 assumptions · 0 invented entities

The central claim rests on three domain assumptions that are plausible but not verified in the small-data regime. The hand-chosen alpha and beta parameters are not covered by the theoretical derivation, and the reward temperature is unreported.

free parameters (3)
  • alpha (acquisition trade-off) = alpha=1 for imputation; alpha=0.4 for active prediction
    Controls the mixture of the imputation reward R_I and the prediction reward R_P in Eq. 10. The information-theoretic equivalence to mutual information is shown for alpha=1/2 only, so the experimental choices are not justified by the stated theory.
  • beta (conditional BELGAM trade-off) = beta=0.6
    Balances the conditional prediction likelihood against the joint reconstruction objective in Eq. 15. No tuning procedure or sensitivity analysis is reported.
  • Temperature T = not specified
    Used in Eq. 28 to convert acquisition rewards into sampling probabilities. The value controls exploration versus exploitation and is never reported, which limits reproducibility.
assumptions (3)
  • domain assumption The amortized inference family F is large enough that the ELBO maximized over q_phi gives an accurate approximation to log p(XO, theta) and its gradient.
    Invoked in Section 2.2, Eq. 3. If the encoder cannot represent the true posterior p(z|x, theta), the SGHMC sampler targets a biased distribution and the uncertainty estimates used for acquisition are miscalibrated.
  • domain assumption Observing a single feature element x_i,d does not change the posterior of the global weights theta.
    Stated in Section 4.2 and used in the derivation of R_P and the EDDI approximation, where KL terms involving theta updates are dropped. Plausible for large training sets but questionable in the ice-start regime.
  • domain assumption Features are conditionally independent given the latent variable z and the weights theta.
    Used in Appendix C.1, Eq. 20, to factor the joint distribution and derive the EDDI reward for PA-BELGAM.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Icebreaker: Element-wise Active Information Acquisition with Bayesian Deep Latent Gaussian Model." pith.science (2026). https://pith.science/paper/3SGF5QEZ

@misc{pith2026190804537,
  author       = {Pith},
  title        = {Pith review of: Icebreaker: Element-wise Active Information Acquisition with Bayesian Deep Latent Gaussian Model},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/3SGF5QEZ}},
  note         = {Machine review of arXiv:1908.04537}
}
read the original abstract

In this paper we introduce the ice-start problem, i.e., the challenge of deploying machine learning models when only little or no training data is initially available, and acquiring each feature element of data is associated with costs. This setting is representative for the real-world machine learning applications. For instance, in the health-care domain, when training an AI system for predicting patient metrics from lab tests, obtaining every single measurement comes with a high cost. Active learning, where only the label is associated with a cost does not apply to such problem, because performing all possible lab tests to acquire a new training datum would be costly, as well as unnecessary due to redundancy. We propose Icebreaker, a principled framework to approach the ice-start problem. Icebreaker uses a full Bayesian Deep Latent Gaussian Model (BELGAM) with a novel inference method. Our proposed method combines recent advances in amortized inference and stochastic gradient MCMC to enable fast and accurate posterior inference. By utilizing BELGAM's ability to fully quantify model uncertainty, we also propose two information acquisition functions for imputation and active prediction problems. We demonstrate that BELGAM performs significantly better than the previous VAE (Variational autoencoder) based models, when the data set size is small, using both machine learning benchmarks and real-world recommender systems and health-care applications. Moreover, based on BELGAM, Icebreaker further improves the performance and demonstrate the ability to use minimum amount of the training data to obtain the highest test time performance.

Figures

Figures reproduced from arXiv: 1908.04537 by the authors.

Figure 1
Figure 1. BELGAM We design a flexible full Bayesian model which provides the model uncertainty quantification. A Bayesian latent variable generative model as shown in [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. The illustration of P-VAE inference network structure. However, in our problem setting, the feature values for each data instance are partially observed. Thus, the vanilla amortized inference cannot be used as the input dimensionality to the network can vary for each data instance. As with the Partial VAE proposed in Ma et al. (2019), we adopt the set encoding structure (Qi et al., 2017; Zaheer et al., 2017) to buil… view at source ↗
Figure 3
Figure 3. The flow diagram of active training phase at time [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (7 more)
Figure 4
Figure 4. Figure 4: Boston Housing experimental results. (a) The NLL over the number of observed [PITH_FULL_IMAGE:figures/full_fig_p010_4.png]
Figure 5
Figure 5. Figure 5: Evaluation of test time performance after exposure to different number of training [PITH_FULL_IMAGE:figures/full_fig_p011_5.png]
Figure 6
Figure 6. Figure 6: Performance using MovieLens. Panel (a) shows the imputation NLL with number [PITH_FULL_IMAGE:figures/full_fig_p011_6.png]
Figure 7
Figure 7. Figure 7: Performance of MIMIC experiments. (Left) This figure shows the predictive AUIC curve as training data size increases. (Middle) The accumulated feature statistics as active selection progresses (Right) This indicates the histogram of initial choice during active predict…
Figure 8
Figure 8. Figure 8: (Left) The missing proportion of each feature in MIMIC III. (Middle and Right) [PITH_FULL_IMAGE:figures/full_fig_p026_8.png]
Figure 10
Figure 10. Figure 10: AUIC curve w.r.t. Energy data set size small. Figure 9b shows the selection pattern of Icebreaker, where we can observe a long-tailed strategy similar to Boston housing and MovieLens-1m imputation. Active Prediction [PITH_FULL_IMAGE:figures/full_fig_p027_10.png]
Figure 11
Figure 11. Figure 11: (Left) Accumulated feature number (Middle) Correlations between the features [PITH_FULL_IMAGE:figures/full_fig_p028_11.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

58 extracted references · 49 canonical work pages

  1. [1]

    An introduction to MCMC for machine learning

    Christophe Andrieu, Nando De Freitas, Arnaud Doucet, and Michael I Jordan. An introduction to MCMC for machine learning. Machine learning, 50 0 (1-2): 0 5--43, 2003

    work page 2003

  2. [2]

    Online choice of active learning algorithms

    Yoram Baram, Ran El Yaniv, and Kobi Luz. Online choice of active learning algorithms. Journal of Machine Learning Research, 5 0 (Mar): 0 255--291, 2004

    work page 2004

  3. [3]

    Variational algorithms for approximate Bayesian inference

    Matthew James Beal et al. Variational algorithms for approximate Bayesian inference. 2003

    work page 2003

  4. [4]

    Expected information as expected utility

    Jos \'e M Bernardo. Expected information as expected utility. The Annals of Statistics, pages 686--690, 1979

    work page 1979

  5. [5]

    Active matrix completion

    Shayok Chakraborty, Jiayu Zhou, Vineeth Balasubramanian, Sethuraman Panchanathan, Ian Davidson, and Jieping Ye. Active matrix completion. In 2013 IEEE 13th International Conference on Data Mining, pages 81--90. IEEE, 2013

    work page 2013

  6. [6]

    Bridging the gap between stochastic gradient MCMC and stochastic optimization

    Changyou Chen, David Carlson, Zhe Gan, Chunyuan Li, and Lawrence Carin. Bridging the gap between stochastic gradient MCMC and stochastic optimization. In Artificial Intelligence and Statistics, pages 1051--1060, 2016

    work page 2016

  7. [7]

    Elements of information theory

    Thomas M Cover and Joy A Thomas. Elements of information theory. John Wiley & Sons, 2012

    2012

  8. [8]

    UCI machine learning repository, 2017

    Dua Dheeru and Efi Karra Taniskidou. UCI machine learning repository, 2017. URL http://archive.ics.uci.edu/ml

    work page 2017

Show all 58 references
  1. [9]

    Deep bayesian active learning with image data

    Yarin Gal, Riashat Islam, and Zoubin Ghahramani. Deep bayesian active learning with image data. In Proceedings of the 34th International Conference on Machine Learning-Volume 70, pages 1183--1192. JMLR. org, 2017

    2017

  2. [10]

    The Movielens datasets: History and context

    F Maxwell Harper and Joseph A Konstan. The Movielens datasets: History and context. Acm transactions on interactive intelligent systems (tiis), 5 0 (4): 0 19, 2016

    2016

  3. [11]

    Multitask learning and benchmarking with clinical time series data

    Hrayr Harutyunyan, Hrant Khachatrian, David C Kale, and Aram Galstyan. Multitask learning and benchmarking with clinical time series data. arXiv preprint arXiv:1703.07771, 2017

    2017 arXiv

  4. [12]

    Inference in deep gaussian processes using stochastic gradient Hamiltonian Monte Carlo

    Marton Havasi, Jos \'e Miguel Hern \'a ndez-Lobato, and Juan Jos \'e Murillo-Fuentes. Inference in deep gaussian processes using stochastic gradient Hamiltonian Monte Carlo . In Advances in Neural Information Processing Systems, pages 7506--7516, 2018

    2018

  5. [13]

    Bayesian active learning for classification and preference learning

    Neil Houlsby, Ferenc Husz \'a r, Zoubin Ghahramani, and M \'a t \'e Lengyel. Bayesian active learning for classification and preference learning. arXiv preprint arXiv:1112.5745, 2011

    2011 arXiv

  6. [14]

    Cold-start active learning with robust ordinal matrix factorization

    Neil Houlsby, Jos \'e Miguel Hern \'a ndez-Lobato, and Zoubin Ghahramani. Cold-start active learning with robust ordinal matrix factorization. In International Conference on Machine Learning, pages 766--774, 2014

    2014

  7. [15]

    Active feature acquisition with supervised matrix completion

    Sheng-Jun Huang, Miao Xu, Ming-Kun Xie, Masashi Sugiyama, Gang Niu, and Songcan Chen. Active feature acquisition with supervised matrix completion. arXiv preprint arXiv:1802.05380, 2018

    2018 arXiv

  8. [16]

    Classification with costly features using deep reinforcement learning

    Jarom \' r Janisch, Tom \'a s Pevn \`y , and Viliam Lis \`y . Classification with costly features using deep reinforcement learning. arXiv preprint arXiv:1711.07364, 2017

    2017 arXiv

  9. [17]

    MIMIC-III , a freely accessible critical care database

    Alistair EW Johnson, Tom J Pollard, Lu Shen, H Lehman Li-wei, Mengling Feng, Mohammad Ghassemi, Benjamin Moody, Peter Szolovits, Leo Anthony Celi, and Roger G Mark. MIMIC-III , a freely accessible critical care database. Scientific Data, 3: 0 160035, 2016

    2016

  10. [18]

    An introduction to variational methods for graphical models

    Michael I Jordan, Zoubin Ghahramani, Tommi S Jaakkola, and Lawrence K Saul. An introduction to variational methods for graphical models. Machine learning, 37 0 (2): 0 183--233, 1999

    1999

  11. [19]

    Selective supervision: Guiding supervised learning with decision-theoretic active learning

    Ashish Kapoor, Eric Horvitz, and Sumit Basu. Selective supervision: Guiding supervised learning with decision-theoretic active learning

  12. [20]

    Auto-encoding variational Bayes

    Diederik P Kingma and Max Welling. Auto-encoding variational Bayes . In International Conference on Learning Representation, 2014

    2014

  13. [21]

    A utility-theoretic approach to privacy in online services

    Andreas Krause and Eric Horvitz. A utility-theoretic approach to privacy in online services. Journal of Artificial Intelligence Research, 39: 0 633--662, 2010

    2010

  14. [22]

    Traffic updates: Saying a lot while revealing a little

    John Krumm and Eric Horvitz. Traffic updates: Saying a lot while revealing a little. 2019

    2019

  15. [23]

    Knowing what to ask: A Bayesian active learning approach to the surveying problem

    Yoad Lewenberg, Yoram Bachrach, Ulrich Paquet, and Jeffrey S Rosenschein. Knowing what to ask: A Bayesian active learning approach to the surveying problem. In AAAI, pages 1396--1402, 2017

    2017

  16. [24]

    Preconditioned stochastic gradient Langevin dynamics for deep neural networks

    Chunyuan Li, Changyou Chen, David Carlson, and Lawrence Carin. Preconditioned stochastic gradient Langevin dynamics for deep neural networks. In Thirtieth AAAI Conference on Artificial Intelligence, 2016

    2016

  17. [25]

    Approximate Inference: New Visions

    Yingzhen Li. Approximate Inference: New Visions. PhD thesis, University of Cambridge, 2018

    2018

  18. [26]

    On a measure of the information provided by an experiment

    Dennis V Lindley. On a measure of the information provided by an experiment. The Annals of Mathematical Statistics, pages 986--1005, 1956

    1956

  19. [27]

    Partial VAE for hybrid recommender system

    Chao Ma, Wenbo Gong, Jos \'e Miguel Hern \'a ndez-Lobato, Noam Koenigstein, Sebastian Nowozin, and Cheng Zhang. Partial VAE for hybrid recommender system. In NIPS Workshop on Bayesian Deep Learning, 2018

    2018

  20. [28]

    EDDI : Efficient dynamic discovery of high-value information with partial vae

    Chao Ma, Sebastian Tschiatschek, Konstantina Palla, Jose Miguel Hernandez Lobato, Sebastian Nowozin, and Cheng Zhang. EDDI : Efficient dynamic discovery of high-value information with partial vae. In Proceedings of the International Conference on Machine Learning, 2019

    2019

  21. [29]

    Information-based objective functions for active data selection

    David JC MacKay. Information-based objective functions for active data selection. Neural computation, 4 0 (4): 0 590--604, 1992

    1992

  22. [30]

    Pointing the way: active collaborative filtering

    David Maltz and Kate Ehrlich. Pointing the way: active collaborative filtering

  23. [31]

    Employing EM and pool-based active learning for text classification

    Andrew Kachites McCallumzy and Kamal Nigamy. Employing EM and pool-based active learning for text classification. In International Conference on Machine Learning, pages 359--367. Citeseer, 1998

    1998

  24. [32]

    Active feature-value acquisition for classifier induction

    Prem Melville, Maytal Saar-Tsechansky, Foster Provost, and Raymond Mooney. Active feature-value acquisition for classifier induction. In International Conference on Data Mining, pages 483--486. IEEE, 2004

    2004

  25. [33]

    An expected utility approach to active feature-value acquisition

    Prem Melville, Maytal Saar-Tsechansky, Foster Provost, and Raymond Mooney. An expected utility approach to active feature-value acquisition. In Fifth IEEE International Conference on Data Mining (ICDM'05), pages 4--pp. IEEE, 2005

    2005

  26. [34]

    Handling incomplete heterogeneous data using VAEs

    Alfredo Nazabal, Pablo M Olmos, Zoubin Ghahramani, and Isabel Valera. Handling incomplete heterogeneous data using VAEs . arXiv preprint arXiv:1807.03653, 2018

    2018 arXiv

  27. [35]

    Resolving cold start problem in recommendation system using demographic approach

    Anand Kishor Pandey and Dharmveer Singh Rajpoot. Resolving cold start problem in recommendation system using demographic approach. In 2016 International Conference on Signal Processing and Communication (ICSC), pages 213--218. IEEE, 2016

    2016

  28. [36]

    Interpretable outcome prediction with sparse Bayesian neural networks in intensive care

    Anna-Lena Popkes, Hiske Overweg, Ari Ercole, Yingzhen Li, Jos \'e Miguel Hern \'a ndez-Lobato, Yordan Zaykov, and Cheng Zhang. Interpretable outcome prediction with sparse Bayesian neural networks in intensive care. arXiv preprint arXiv:1905.02599, 2019

    1905 arXiv

  29. [37]

    Pointnet: Deep learning on point sets for 3D classification and segmentation

    Charles R Qi, Hao Su, Kaichun Mo, and Leonidas J Guibas. Pointnet: Deep learning on point sets for 3D classification and segmentation. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 652--660, 2017

    2017

  30. [38]

    Stochastic backpropagation and approximate inference in deep generative models

    Danilo Jimenez Rezende, Shakir Mohamed, and Daan Wierstra. Stochastic backpropagation and approximate inference in deep generative models. In Interantional Conference on Machine Learning, 2014

    2014

  31. [39]

    Toward optimal active learning through monte carlo estimation of error reduction

    Nicholas Roy and Andrew McCallum. Toward optimal active learning through monte carlo estimation of error reduction

  32. [40]

    Matrix completion with queries

    Natali Ruchansky, Mark Crovella, and Evimaria Terzi. Matrix completion with queries. In Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pages 1025--1034. ACM, 2015

    2015

  33. [41]

    Active feature-value acquisition

    Maytal Saar-Tsechansky, Prem Melville, and Foster Provost. Active feature-value acquisition. Management Science, 55 0 (4): 0 664--684, 2009

    2009

  34. [42]

    Methods and metrics for cold-start recommendations

    Andrew I Schein, Alexandrin Popescul, Lyle H Ungar, and David M Pennock. Methods and metrics for cold-start recommendations. In Proceedings of the 25th annual international ACM SIGIR conference on Research and development in information retrieval, pages 253--260. ACM, 2002

    2002

  35. [43]

    Active learning

    Burr Settles. Active learning. Synthesis Lectures on Artificial Intelligence and Machine Learning, 6 0 (1): 0 1--114, 2012

    2012

  36. [44]

    Joint active feature acquisition and classification with variable-size set encoding

    Hajin Shim, Sung Ju Hwang, and Eunho Yang. Joint active feature acquisition and classification with variable-size set encoding. In Advances in Neural Information Processing Systems, 2018

    2018

  37. [45]

    Learning structured output representation using deep conditional generative models

    Kihyuk Sohn, Honglak Lee, and Xinchen Yan. Learning structured output representation using deep conditional generative models. In Advances in neural information processing systems, pages 3483--3491, 2015

    2015

  38. [46]

    Predicting outcome after traumatic brain injury: development and international validation of prognostic scores based on admission characteristics

    Ewout W Steyerberg, Nino Mushkudiani, Pablo Perel, Isabella Butcher, Juan Lu, Gillian S McHugh, Gordon D Murray, Anthony Marmarou, Ian Roberts, J Dik F Habbema, et al. Predicting outcome after traumatic brain injury: development and international validation of prognostic score...

    2008

  39. [47]

    Active learning and search on low-rank matrices

    Dougal J Sutherland, Barnab \'a s P \'o czos, and Jeff Schneider. Active learning and search on low-rank matrices. In Proceedings of the 19th ACM SIGKDD international conference on Knowledge discovery and data mining, pages 212--220. ACM, 2013

    2013

  40. [48]

    An efficient heuristic method for active feature acquisition and its application to protein-protein interaction prediction

    Mohamed Thahir, Tarun Sharma, and Madhavi K Ganapathiraju. An efficient heuristic method for active feature acquisition and its application to protein-protein interaction prediction. In BMC proceedings, volume 6, page S2. BioMed Central, 2012

    2012

  41. [49]

    Support vector machine active learning with applications to text classification

    Simon Tong and Daphne Koller. Support vector machine active learning with applications to text classification. Journal of machine learning research, 2 0 (Nov): 0 45--66, 2001

    2001

  42. [50]

    Intelligent information acquisition for improved clustering

    Duy Vu, Prem Melville, Mikhail Bilenko, and Maytal Saar-Tsechansky. Intelligent information acquisition for improved clustering

  43. [51]

    Graphical models, exponential families, and variational inference

    Martin J Wainwright, Michael I Jordan, et al. Graphical models, exponential families, and variational inference. Foundations and Trends in Machine Learning , 1 0 (1--2): 0 1--305, 2008

    2008

  44. [52]

    A Monte Carlo implementation of the EM algorithm and the poor man's data augmentation algorithms

    Greg CG Wei and Martin A Tanner. A Monte Carlo implementation of the EM algorithm and the poor man's data augmentation algorithms. Journal of the American statistical Association, 85 0 (411): 0 699--704, 1990

    1990

  45. [53]

    Ice-breaking: mitigating cold-start recommendation problem by rating comparison

    Jingwei Xu, Yuan Yao, Hanghang Tong, Xianping Tao, and Jian Lu. Ice-breaking: mitigating cold-start recommendation problem by rating comparison. In Twenty-Fourth International Joint Conference on Artificial Intelligence, 2015

    2015

  46. [54]

    Deep sets

    Manzil Zaheer, Satwik Kottur, Siamak Ravanbakhsh, Barnabas Poczos, Ruslan R Salakhutdinov, and Alexander J Smola. Deep sets. In Advances in Neural Information Processing Systems, pages 3391--3401, 2017

    2017

  47. [55]

    Odin: Optimal discovery of high-value information using model-based deep reinforcement learning

    Sara Zannone, Jos \'e Miguel Hern \'a ndez-Lobato, Cheng Zhang, and Konstantina Palla. Odin: Optimal discovery of high-value information using model-based deep reinforcement learning. In ICML Real-world Sequential Decision Making Workshop, 2019

    2019

  48. [56]

    Advances in variational inference

    Cheng Zhang, Judith Butepage, Hedvig Kjellstrom, and Stephan Mandt. Advances in variational inference. IEEE transactions on pattern analysis and machine intelligence, 2018

    2018

  49. [57]

    Generative adversarial active learning

    Jia-Jie Zhu and Jos \'e Bento. Generative adversarial active learning. arXiv preprint arXiv:1702.07956, 2017

    2017 arXiv

  50. [58]

    Combining active learning and semi-supervised learning using gaussian fields and harmonic functions

    Xiaojin Zhu, John Lafferty, and Zoubin Ghahramani. Combining active learning and semi-supervised learning using gaussian fields and harmonic functions

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.