We study the optimization landscape of the log-likelihood function and the convergence of the Expectation-Maximization (EM) algorithm in latent Gaussian tree models, i.e. tree-structured Gaussian graphical models whose leaf nodes are observable and non-leaf nodes are unobservable. We show that the unique non-trivial stationary point of the population log-likelihood is its global maximum, and establish that the expectation-maximization algorithm is guaranteed to converge to it in the single latent variable case. Our results for the landscape of the log-likelihood function in general latent tree models provide support for the extensive practical use of maximum likelihood based-methods in this setting. Our results for the EM algorithm extend an emerging line of work on obtaining global convergence guarantees for this celebrated algorithm. We show our results for the non-trivial stationary points of the log-likelihood by arguing that a certain system of polynomial equations obtained from the EM updates has a unique non-trivial solution. The global convergence of the EM algorithm follows by arguing that all trivial fixed points are higher-order saddle points.

Private multi-winner voting is the task of revealing $k$-hot binary vectors satisfying a bounded differential privacy (DP) guarantee. This task has been understudied in machine learning literature despite its prevalence in many domains such as healthcare. We propose three new DP multi-winner mechanisms: Binary, $\tau$, and Powerset voting. Binary voting operates independently per label through composition. $\tau$ voting bounds votes optimally in their $\ell_2$ norm for tight data-independent guarantees. Powerset voting operates over the entire binary vector by viewing the possible outcomes as a power set. Our theoretical and empirical analysis shows that Binary voting can be a competitive mechanism on many tasks unless there are strong correlations between labels, in which case Powerset voting outperforms it. We use our mechanisms to enable privacy-preserving multi-label learning in the central setting by extending the canonical single-label technique: PATE. We find that our techniques outperform current state-of-the-art approaches on large, real-world healthcare data and standard multi-label benchmarks. We further enable multi-label confidential and private collaborative (CaPC) learning and show that model performance can be significantly improved in the multi-site setting.

The International Workshop on Reading Music Systems (WoRMS) is a workshop that tries to connect researchers who develop systems for reading music, such as in the field of Optical Music Recognition, with other researchers and practitioners that could benefit from such systems, like librarians or musicologists. The relevant topics of interest for the workshop include, but are not limited to: Music reading systems; Optical music recognition; Datasets and performance evaluation; Image processing on music scores; Writer identification; Authoring, editing, storing and presentation systems for music scores; Multi-modal systems; Novel input-methods for music to produce written music; Web-based Music Information Retrieval services; Applications and projects; Use-cases related to written music. These are the proceedings of the 4th International Workshop on Reading Music Systems, held online on Nov. 18th 2022.

Unsupervised skill learning aims to learn a rich repertoire of behaviors without external supervision, providing artificial agents with the ability to control and influence the environment. However, without appropriate knowledge and exploration, skills may provide control only over a restricted area of the environment, limiting their applicability. Furthermore, it is unclear how to leverage the learned skill behaviors for adapting to downstream tasks in a data-efficient manner. We present Choreographer, a model-based agent that exploits its world model to learn and adapt skills in imagination. Our method decouples the exploration and skill learning processes, being able to discover skills in the latent state space of the model. During adaptation, the agent uses a meta-controller to evaluate and adapt the learned skills efficiently by deploying them in parallel in imagination. Choreographer is able to learn skills both from offline data, and by collecting data simultaneously with an exploration policy. The skills can be used to effectively adapt to downstream tasks, as we show in the URL benchmark, where we outperform previous approaches from both pixels and states inputs. The learned skills also explore the environment thoroughly, finding sparse rewards more frequently, as shown in goal-reaching tasks from the DMC Suite and Meta-World.

In maximum-likelihood quantum state tomography, both the sample size and dimension grow exponentially with the number of qubits. It is therefore desirable to develop a stochastic first-order method, just like stochastic gradient descent for modern machine learning, to compute the maximum-likelihood estimate. To this end, we propose an algorithm called stochastic mirror descent with the Burg entropy. Its expected optimization error vanishes at a $O ( \sqrt{ ( 1 / t ) d \log t } )$ rate, where $d$ and $t$ denote the dimension and number of iterations, respectively. Its per-iteration time complexity is $O ( d^3 )$, independent of the sample size. To the best of our knowledge, this is currently the computationally fastest stochastic first-order method for maximum-likelihood quantum state tomography.

Internet of Things (IoT) networks have become an increasingly attractive target of cyberattacks. Powerful Machine Learning (ML) models have recently been adopted to implement network intrusion detection systems to protect IoT networks. For the successful training of such ML models, selecting the right data features is crucial, maximising the detection accuracy and computational efficiency. This paper comprehensively analyses feature sets' importance and predictive power for detecting network attacks. Three feature selection algorithms: chi-square, information gain and correlation, have been utilised to identify and rank data features. The attributes are fed into two ML classifiers: deep feed-forward and random forest, to measure their attack detection performance. The experimental evaluation considered three datasets: UNSW-NB15, CSE-CIC-IDS2018, and ToN-IoT in their proprietary flow format. In addition, the respective variants in NetFlow format were also considered, i.e., NF-UNSW-NB15, NF-CSE-CIC-IDS2018, and NF-ToN-IoT. The experimental evaluation explored the marginal benefit of adding individual features. Our results show that the accuracy initially increases rapidly with adding features but converges quickly to the maximum. This demonstrates a significant potential to reduce the computational and storage cost of intrusion detection systems while maintaining near-optimal detection accuracy. This has particular relevance in IoT systems, with typically limited computational and storage resources.

This paper presents a new perspective of self-supervised learning based on extending heat equation into high dimensional feature space. In particular, we remove time dependence by steady-state condition, and extend the remaining 2D Laplacian from x--y isotropic to linear correlated. Furthermore, we simplify it by splitting x and y axes as two first-order linear differential equations. Such simplification explicitly models the spatial invariance along horizontal and vertical directions separately, supporting prediction across image blocks. This introduces a very simple masked image modeling (MIM) method, named QB-Heat. QB-Heat leaves a single block with size of quarter image unmasked and extrapolates other three masked quarters linearly. It brings MIM to CNNs without bells and whistles, and even works well for pre-training light-weight networks that are suitable for both image classification and object detection without fine-tuning. Compared with MoCo-v2 on pre-training a Mobile-Former with 5.8M parameters and 285M FLOPs, QB-Heat is on par in linear probing on ImageNet, but clearly outperforms in non-linear probing that adds a transformer block before linear classifier (65.6% vs. 52.9%). When transferring to object detection with frozen backbone, QB-Heat outperforms MoCo-v2 and supervised pre-training on ImageNet by 7.9 and 4.5 AP respectively. This work provides an insightful hypothesis on the invariance within visual representation over different shapes and textures: the linear relationship between horizontal and vertical derivatives. The code will be publicly released.

Big data analysis poses the dual problem of privacy preservation and utility, i.e., how accurate data analyses remain after transforming original data in order to protect the privacy of the individuals that the data is about - and whether they are accurate enough to be meaningful. In this paper, we thus investigate across several datasets whether different methods of generating fully synthetic data vary in their utility a priori (when the specific analyses to be performed on the data are not known yet), how closely their results conform to analyses on original data a posteriori, and whether these two effects are correlated. We find some methods (decision-tree based) to perform better than others across the board, sizeable effects of some choices of imputation parameters (notably the number of released datasets), no correlation between broad utility metrics and analysis accuracy, and varying correlations for narrow metrics. We did get promising findings for classification tasks when using synthetic data for training machine learning models, which we consider worth exploring further also in terms of mitigating privacy attacks against ML models such as membership inference and model inversion.

Precision measurements and new physics searches at the Large Hadron Collider require efficient simulations of particle propagation and interactions within the detectors. The most computationally expensive simulations involve calorimeter showers. Advances in deep generative modelling - particularly in the realm of high-dimensional data - have opened the possibility of generating realistic calorimeter showers orders of magnitude more quickly than physics-based simulation. However, the high-dimensional representation of showers belies the relative simplicity and structure of the underlying physical laws. This phenomenon is yet another example of the manifold hypothesis from machine learning, which states that high-dimensional data is supported on low-dimensional manifolds. We thus propose modelling calorimeter showers first by learning their manifold structure, and then estimating the density of data across this manifold. Learning manifold structure reduces the dimensionality of the data, which enables fast training and generation when compared with competing methods.

An open problem on the path to artificial intelligence is generalization from the known to the unknown, which is instantiated as Generalized Zero-Shot Learning (GZSL) task. In this work, we propose a novel Evolutionary Generalized Zero-Shot Learning setting, which (i) avoids the domain shift problem in inductive GZSL, and (ii) is more in line with the needs of real-world deployments than transductive GZSL. In the proposed setting, a zero-shot model with poor initial performance is able to achieve online evolution during application. We elaborate on three challenges of this special task, i.e., catastrophic forgetting, initial prediction bias, and evolutionary data class bias. Moreover, we propose targeted solutions for each challenge, resulting in a generic method capable of continuing to evolve on a given initial IGZSL model. Experiments on three popular GZSL benchmark datasets show that our model can learn from the test data stream while other baselines fail.