Deep learning faces significant challenges during the training of neural networks, including internal covariate shift, label shift, vanishing/exploding gradients, overfitting, and computational complexity. While conventional normalization methods, such as Batch Normalization, aim to tackle some of these issues, they often depend on assumptions that constrain their adaptability. Mixture Normalization faces computational hurdles in its pursuit of handling multiple Gaussian distributions. This paper introduces Cluster-Based Normalization (CB-Norm) in two variants - Supervised Cluster-Based Normalization (SCB-Norm) and Unsupervised Cluster-Based Normalization (UCB-Norm) - proposing a groundbreaking one-step normalization approach. CB-Norm leverages a Gaussian mixture model to specifically address challenges related to gradient stability and learning acceleration. For SCB-Norm, a supervised variant, the novel mechanism involves introducing predefined data partitioning, termed clusters, to normalize activations based on the assigned cluster. This cluster-driven approach creates a space that conforms to a Gaussian mixture model. On the other hand, UCB-Norm, an unsupervised counterpart, dynamically clusters neuron activations during training, adapting to task-specific challenges without relying on predefined data partitions (clusters). This dual approach ensures flexibility in addressing diverse learning scenarios. CB-Norm innovatively uses a one-step normalization approach, where parameters of each mixture component (cluster in activation space) serve as weights for deep neural networks. This adaptive clustering process tackles both clustering and resolution of deep neural network tasks concurrently during training, signifying a notable advancement in the field.
The fusion models, which effectively combine information from different sources, are widely used in solving multimodal tasks. However, they have significant limitations related to aligning data distributions across different modalities. This challenge can lead to inconsistencies and difficulties in learning robust representations. Alignment models, while specifically addressing this issue, often require training "from scratch" with large datasets to achieve optimal results, which can be costly in terms of resources and time. To overcome these limitations, we propose an innovative model called Context-Based Multimodal Fusion (CBMF), which combines both modality fusion and data distribution alignment. In CBMF, each modality is represented by a specific context vector, fused with the embedding of each modality. This enables the use of large pre-trained models that can be frozen, reducing the computational and training data requirements. Additionally, the network learns to differentiate embeddings of different modalities through fusion with context and aligns data distributions using a contrastive approach for self-supervised learning. Thus, CBMF offers an effective and economical solution for solving complex multimodal tasks.
In this paper, we introduce a novel Distributed Markov Chain Monte Carlo (MCMC) inference method for the Bayesian Non-Parametric Latent Block Model (DisNPLBM), employing the Master/Worker architecture. Our non-parametric co-clustering algorithm divides observations and features into partitions using latent multivariate Gaussian block distributions. The workload on rows is evenly distributed among workers, who exclusively communicate with the master and not among themselves. DisNPLBM demonstrates its impact on cluster labeling accuracy and execution times through experimental results. Moreover, we present a real-use case applying our approach to co-cluster gene expression data. The code source is publicly available at https://github.com/redakhoufache/Distributed-NPLBM.
Dirichlet Process Mixture Models (DPMMs) are widely used to address clustering problems. Their main advantage lies in their ability to automatically estimate the number of clusters during the inference process through the Bayesian non-parametric framework. However, the inference becomes considerably slow as the dataset size increases. This paper proposes a new distributed Markov Chain Monte Carlo (MCMC) inference method for DPMMs (DisCGS) using sufficient statistics. Our approach uses the collapsed Gibbs sampler and is specifically designed to work on distributed data across independent and heterogeneous machines, which habilitates its use in horizontal federated learning. Our method achieves highly promising results and notable scalability. For instance, with a dataset of 100K data points, the centralized algorithm requires approximately 12 hours to complete 100 iterations while our approach achieves the same number of iterations in just 3 minutes, reducing the execution time by a factor of 200 without compromising clustering performance. The code source is publicly available at https://github.com/redakhoufache/DisCGS.
Machine Learning approaches like clustering methods deal with massive datasets that present an increasing challenge. We devise parallel algorithms to compute the Multi-Slice Clustering (MSC) for 3rd-order tensors. The MSC method is based on spectral analysis of the tensor slices and works independently on each tensor mode. Such features fit well in the parallel paradigm via a distributed memory system. We show that our parallel scheme outperforms sequential computing and allows for the scalability of the MSC method.
Apart from the high accuracy of machine learning models, what interests many researchers in real-life problems (e.g., fraud detection, credit scoring) is to find hidden patterns in data; particularly when dealing with their challenging imbalanced characteristics. Interpretability is also a key requirement that needs to accompany the used machine learning model. In this concern, often, intrinsically interpretable models are preferred to complex ones, which are in most cases black-box models. Also, linear models are used in some high-risk fields to handle tabular data, even if performance must be sacrificed. In this paper, we introduce Self-Reinforcement Attention (SRA), a novel attention mechanism that provides a relevance of features as a weight vector which is used to learn an intelligible representation. This weight is then used to reinforce or reduce some components of the raw input through element-wise vector multiplication. Our results on synthetic and real-world imbalanced data show that our proposed SRA block is effective in end-to-end combination with baseline models.
Several methods for triclustering three-dimensional data require the cluster size or the number of clusters in each dimension to be specified. To address this issue, the Multi-Slice Clustering (MSC) for 3-order tensor finds signal slices that lie in a low dimensional subspace for a rank-one tensor dataset in order to find a cluster based on the threshold similarity. We propose an extension algorithm called MSC-DBSCAN to extract the different clusters of slices that lie in the different subspaces from the data if the dataset is a sum of r rank-one tensor (r > 1). Our algorithm uses the same input as the MSC algorithm and can find the same solution for rank-one tensor data as MSC.
Genetic algorithms are a well-known example of bio-inspired heuristic methods. They mimic natural selection by modeling several operators such as mutation, crossover, and selection. Recent discoveries about Epigenetics regulation processes that occur "on top of" or "in addition to" the genetic basis for inheritance involve changes that affect and improve gene expression. They raise the question of improving genetic algorithms (GAs) by modeling epigenetics operators. This paper proposes a new epigenetics algorithm that mimics the epigenetics phenomenon known as DNA methylation. The novelty of our epigenetics algorithms lies primarily in taking advantage of attention mechanisms and deep learning, which fits well with the genes enhancing/silencing concept. The paper develops theoretical arguments and presents empirical studies to exhibit the capability of the proposed epigenetics algorithms to solve more complex problems efficiently than has been possible with simple GAs; for example, facing two Non-convex (multi-peaks) optimization problems as presented in this paper, the proposed epigenetics algorithm provides good performances and shows an excellent ability to overcome the lack of local optimum and thus find the global optimum.
Several methods for triclustering three-dimensional data require the cluster size or the number of clusters in each dimension to be specified. To address this issue, the Multi-Slice Clustering (MSC) for 3-order tensor finds signal slices that lie in a low dimensional subspace for a rank-one tensor dataset in order to find a cluster based on the threshold similarity. We propose an extension algorithm called MSC-DBSCAN to extract the different clusters of slices that lie in the different subspaces from the data if the dataset is a sum of r rank-one tensor (r > 1). Our algorithm uses the same input as the MSC algorithm and can find the same solution for rank-one tensor data as MSC.