arXiv:2606. 18676v1 Announce Type: new Abstract: Training-free neural architecture search promises efficient discovery of high-performance networks without costly training.
arXiv:2605. 15435v2 Announce Type: replace Abstract: Standard deep-learning pipelines usually choose the network architecture before training and keep it fixed throughout optimization.
ONNX-Net introduces a universal representation for neural architectures using natural language descriptions, enabling instant performance prediction across diverse search spaces. The authors present ONNX-Bench, a benchmark of over 600k architecture–accuracy pairs compiled from open‑source NAS‑bench networks in ONNX format. Experiments demonstrate strong zero‑shot predictive performance with minimal pretraining, overcoming the limitations of cell‑based, graph‑encoded approaches.
The paper introduces Neural Spectral Capacity (NSC), a closed‑form metric derived from the singular‑value spectrum of weight matrices that can be computed solely from a network’s architectural specification. Unlike traditional measures such as #Params and #FLOPs, NSC captures architectural structure (depth, width, head, FFN allocations) and can be evaluated without instantiating the model, data, or gradients. Using a dynamic‑programming solver (NSC‑DP), the authors demonstrate that NSC can efficiently identify architectures that outperform existing training‑free proxies across Transformer and CNN families, and achieve state‑of‑the‑art results in tasks such as WikiText‑103 and commonsense reasoning with LLaMA‑7B. whyItMatters":"NSC provides a fast, architecture‑only proxy that outperforms conventional metrics and training‑free proxies, enabling more effective design and pruning of large models without costly training or data."
The paper introduces a pointwise generalization theory for fully connected deep neural networks, using a pointwise Riemannian Dimension derived from eigenvalues of learned feature representations across layers. This framework provides hypothesis-dependent, representation-aware generalization bounds that are significantly tighter than traditional size- or norm-based approaches, both theoretically and experimentally. The authors analytically identify structural properties that explain deep networks’ tractability and empirically show that the pointwise Riemannian Dimension captures feature compression, over‑parameterization effects, and optimizer bias.
arXiv:2604. 15613v4 Announce Type: replace-cross Abstract: We present Green-ELM, a non-iterative neural architecture that replaces gradient-based optimization of the output layer with a closed-form analytic solution over a fixed, high-dimensional random feature representation.
arXiv:2610.00420v1 Announce Type: new Abstract: A weight space network (or metanetwork) takes the weights of another neural network as input and predicts properties of it. Most prior work trains such...
arXiv:2610.00753v1 Announce Type: cross Abstract: End-to-end backpropagation has been the dominant mode of training in deep learning, allowing for the coordination of parameter updates across layers...
arXiv:2606. 19941v1 Announce Type: new Abstract: Compositionality is believed to be the foundation for generalization, enabling models to reuse meaningful primitives in novel combinations.
arXiv:2607. 09757v1 Announce Type: cross Abstract: Low-Rank Adaptation (LoRA) has become a cornerstone of parameter-efficient fine-tuning (PEFT); however, the conventional practice of uniform rank assignment ignores the functional heterogeneity of neural layers.
arXiv:2606. 30512v1 Announce Type: cross Abstract: Why overparameterised deep networks generalise so remarkably well remains one of the most stubborn open questions in machine learning theory.
arXiv:2606. 04754v1 Announce Type: new Abstract: Many striking phenomena in deep learning, such as linear mode connectivity and the structured behavior of training dynamics, are closely tied to parameter symmetries: transformations that leave the realized function unchanged.
arXiv:2602. 20062v2 Announce Type: replace Abstract: Pretraining and fine-tuning are central stages in modern machine learning systems.