WP.L8: Efficient CNNs for Image Analysis |
| Session Type: Lecture |
| Time: Wednesday, September 25, 14:10 - 15:58 |
| Location: South Lounge (3F) |
| Session Chair: Anastasios Tefas, Aristotle University of Thessaloniki |
| WP.L8.1: OPTIMIZING THE BIT ALLOCATION FOR COMPRESSION OF WEIGHTS AND ACTIVATIONS OF DEEP NEURAL NETWORKS |
| Zhe Wang; Stanford University |
| Jie Lin; Institute for Infocomm Research (I2R) |
| Vijay Chandrasekhar; Institute for Infocomm Research (I2R) |
| Bernd Girod; Stanford University |
| WP.L8.2: ACCELERATING FRAMEWORK FOR SIMULTANEOUS OPTIMIZATION OF MODEL ARCHITECTURES AND TRAINING HYPERPARAMETERS |
| Youngjoon Choi; Samsung SDS |
| Jongwon Choi; Samsung SDS |
| Hankyu Moon; Samsung SDS |
| Jaeyoung Lee; Samsung SDS |
| Jin-Yeop Chang; Samsung SDS |
| WP.L8.3: A FIXED-POINT QUANTIZATION TECHNIQUE FOR CONVOLUTIONAL NEURAL NETWORKS BASED ON WEIGHT SCALING |
| Norbert Mitschke; Karlsruhe Institute of Technology |
| Michael Heizmann; Karlsruhe Institute of Technology |
| Klaus-Henning Noffz; Silicon Software |
| Ralf Wittmann; Silicon Software |
| WP.L8.4: DISCRIMINANT ANALYSIS REGULARIZATION IN LIGHTWEIGHT DEEP CNN MODELS |
| Maria Tzelepi; Aristotle University of Thessaloniki |
| Anastasios Tefas; Aristotle University of Thessaloniki |
| WP.L8.5: AUTONOMOUS CHOICE OF DEEP NEURAL NETWORK PARAMETERS BY A MODIFIED GENERATIVE ADVERSARIAL NETWORK |
| Yantao Lu; Syracuse University |
| Senem Velipasalar; Syracuse University |
| WP.L8.6: TUNABLE CNN COMPRESSION THROUGH DIMENSIONALITY REDUCTION |
| Lucas Fernández Brillet; Université Grenoble Alpes and STMicroelectronics |
| Stéphane Mancini; Université Grenoble Alpes |
| Sébastien Cleyet-Merle; STMicroelectronics |
| Marina Nicolas; STMicroelectronics |