memory requirements
The amount of memory needed to store model parameters and data during training and inference, influencing the feasibility of deploying AI models on various hardware.
- Compositional Neural Network Verification via Assume-Guarantee Reasoning
- Continual Knowledge Adaptation for Reinforcement Learning
- Neural Networks for Learnable and Scalable Influence Estimation of Instruction Fine-Tuning Data
- Private Training Large-scale Models with Efficient DP-SGD
- Streaming Attention Approximation via Discrepancy Theory