Neural Network Optimization

Infinite Limits of Multi-head Transformer Dynamics NeurIPS 2024 Information Extraction with Differentiable Beam Search on Graph RNNs COLING 2024 Initialization is Critical to Whether Transformers Fit Composite Functions by Reasoning or Memorizing NeurIPS 2024 Initialization of Large Language Models via Reparameterization to Mitigate Loss Spikes EMNLP 2024 In Search of a Data Transformation That Accelerates Neural Field Training CVPR 2024 Investigating Acceleration of LLaMA Inference by Enabling Intermediate Layer Decoding via Instruction Tuning with ‘LITE’ NAACL 2024 Investigation into Training Dynamics of Learned Optimizers (Student Abstract) AAAI 2024 IQ-VFI: Implicit Quadratic Motion Estimation for Video Frame Interpolation CVPR 2024 Iterate Averaging in the Quest for Best Test Error JMLR 2024 Jointly-Learned Exit and Inference for a Dynamic Neural Network ICLR 2024 JumpCoder: Go Beyond Autoregressive Coder via Online Modification ACL 2024 Jump to Conclusions: Short-Cutting Transformers with Linear Transformations COLING 2024 Kronecker-Factored Approximate Curvature for Physics-Informed Neural Networks NeurIPS 2024 Large Stepsize Gradient Descent for Non-Homogeneous Two-Layer Networks: Margin Improvement and Fast Optimization NeurIPS 2024 Latent Neural Operator for Solving Forward and Inverse PDE Problems NeurIPS 2024 Layer Collaboration in the Forward-Forward Algorithm AAAI 2024 LayerSkip: Enabling Early Exit Inference and Self-Speculative Decoding ACL 2024 Layer-wise Importance Matters: Less Memory for Better Performance in Parameter-efficient Fine-tuning of Large Language Models EMNLP 2024 Layerwise Proximal Replay: A Proximal Point Method for Online Continual Learning ICML 2024 Layer-Wise Representation Fusion for Compositional Generalization AAAI 2024 Learnability of high-dimensional targets by two-parameter models and gradient flow NeurIPS 2024 Learning-Based Tracking-before-Detect for RF-Based Unconstrained Indoor Human Tracking IJCAI 2024 Learning Discretized Neural Networks under Ricci Flow JMLR 2024 Learning Encodings for Constructive Neural Combinatorial Optimization Needs to Regret AAAI 2024 Learning Equi-angular Representations for Online Continual Learning CVPR 2024