Explores MP3 encoding, emphasizing reducing bits through lossy compression and utilizing psycho-acoustic models for efficient filtering and quantization.
Explores image compression through various approaches like pixel and block level compression, Discrete Cosine Transform, quantization, and entropy coding.
Explores the provable benefits of overparameterization in model compression, emphasizing the efficiency of deep neural networks and the importance of retraining for improved performance.
Presents an all-analog photoelectronic chip for high-speed vision tasks, addressing challenges in classical computation and proposing a hybrid optical-electrical framework.
