Modern neural networks of the transformer family require the practitioner to decide, before training begins, how many attention heads to use, how deep the network should be, and how wide each component should be. These decisions are made without knowledge of the task, producing architectures that are systematically larger than necessary: empirical studies find that a substantial fraction of heads and layers can be removed after training without performance loss. This paper introduces DDCL-INCRT, an architecture that determines its own structure during training. Two complementary ideas are combined. The first, DDCL (Deep Dual Competitive Learning), replaces the feedforward block with a dictionary of learned prototype vectors representing the most informative directions in the data. The prototypes spread apart automatically, driven by the training objective, without explicit regularisation. The second, INCRT (Incremental Transformer), controls the number of heads: starting from one, it adds a new head only when the directional information uncaptured by existing heads exceeds a threshold. The main theoretical finding is that these two mechanisms reinforce each other: each new head amplifies prototype separation, which in turn raises the signal triggering the next addition. At convergence, the network self-organises into a hierarchy of heads ordered by representational granularity. This hierarchical structure is proved to be unique and minimal, the smallest architecture sufficient for the task, under the stated conditions. Formal guarantees of stability, convergence, and pruning safety are established throughout. The architecture is not something one designs. It is something one derives.

Effective communication in multi-agent reinforcement learning (MARL) is critical for success but constrained by bandwidth, yet past approaches have been limited to complex gating mechanisms that only decide \textit{whether} to communicate, not \textit{how precisely}. Learning to optimize message precision at the bit-level is fundamentally harder, as the required discretization step breaks gradient flow. We address this by generalizing Differentiable Discrete Communication Learning (DDCL), a framework for end-to-end optimization of discrete messages. Our primary contribution is an extension of DDCL to support unbounded signals, transforming it into a universal, plug-and-play layer for any MARL architecture. We verify our approach with three key results. First, through a qualitative analysis in a controlled environment, we demonstrate \textit{how} agents learn to dynamically modulate message precision according to the informational needs of the task. Second, we integrate our variant of DDCL into four state-of-the-art MARL algorithms, showing it reduces bandwidth by over an order of magnitude while matching or exceeding task performance. Finally, we provide direct evidence for the \enquote{Bitter Lesson} in MARL communication: a simple Transformer-based policy leveraging DDCL matches the performance of complex, specialized architectures, questioning the necessity of bespoke communication designs.

Invariant-based Contrastive Learning (ICL) methods have achieved impressive performance across various domains. However, the absence of latent space representation for distortion (augmentation)-related information in the latent space makes ICL sub-optimal regarding training efficiency and robustness in downstream tasks. Recent studies suggest that introducing equivariance into Contrastive Learning (CL) can improve overall performance. In this paper, we rethink the roles of augmentation strategies and equivariance in improving CL efficacy. We propose a novel Equivariant-based Contrastive Learning (ECL) framework, CLeVER (Contrastive Learning Via Equivariant Representation), compatible with augmentation strategies of arbitrary complexity for various mainstream CL methods and model frameworks. Experimental results demonstrate that CLeVER effectively extracts and incorporates equivariant information from data, thereby improving the training efficiency and robustness of baseline models in downstream tasks.

The order of training samples can have a significant impact on the performance of a classifier. Curriculum learning is a method of ordering training samples from easy to hard. This paper proposes the novel idea of a curriculum learning approach called Data Distribution-based Curriculum Learning (DDCL). DDCL uses the data distribution of a dataset to build a curriculum based on the order of samples. Two types of scoring methods known as DDCL (Density) and DDCL (Point) are used to score training samples thus determining their training order. DDCL (Density) uses the sample density to assign scores while DDCL (Point) utilises the Euclidean distance for scoring. We evaluate the proposed DDCL approach by conducting experiments on multiple datasets using a neural network, support vector machine and random forest classifier. Evaluation results show that the application of DDCL improves the average classification accuracy for all datasets compared to standard evaluation without any curriculum. Moreover, analysis of the error losses for a single training epoch reveals that convergence is faster when using DDCL over the no curriculum method.

Self-supervised learning is well known for its remarkable performance in representation learning and various downstream computer vision tasks. Recently, Positive-pair-Only Contrastive Learning (POCL) has achieved reliable performance without the need to construct positive-negative training sets. It reduces memory requirements by lessening the dependency on the batch size. The POCL method typically uses a single loss function to extract the distortion invariant representation (DIR) which describes the proximity of positive-pair representations affected by different distortions. This loss function implicitly enables the model to filter out or ignore the distortion variant representation (DVR) affected by different distortions. However, existing POCL methods do not explicitly enforce the disentanglement and exploitation of the actually valuable DVR. In addition, these POCL methods have been observed to be sensitive to augmentation strategies. To address these limitations, we propose a novel POCL framework named Distortion-Disentangled Contrastive Learning (DDCL) and a Distortion-Disentangled Loss (DDL). Our approach is the first to explicitly disentangle and exploit the DVR inside the model and feature stream to improve the overall representation utilization efficiency, robustness and representation ability. Experiments carried out demonstrate the superiority of our framework to Barlow Twins and Simsiam in terms of convergence, representation quality, and robustness on several benchmark datasets.