SiG-DML_L5: Sequential Learning |
| Session Type: Lecture |
| Time: Thursday, September 1, 16:10 - 17:50 |
| Location: Atlantic 3 |
| Session Chair: Goran Kvaščev, University of Belgrade |
| SiG-DML_L5.1: AUTO-WEIGHTED SEQUENTIAL WASSERSTEIN DISTANCE AND APPLICATION TO SEQUENCE MATCHING |
| Mitsuhiko Horie, Hiroyuki Kasai, Waseda University, Japan |
| SiG-DML_L5.2: ARE DEEP LEARNING MODELS PRACTICALLY GOOD AS PROMISED? A STRATEGICAL COMPARISON OF DEEP LEARNING MODELS FOR TIME SERIES FORECASTING |
| Zuokun OUYANG, Philippe RAVIER, Meryem JABLOUN, University of Orléans, France |
| SiG-DML_L5.3: CAMEO: CURIOSITY AUGMENTED METROPOLIS FOR EXPLORATORY OPTIMAL POLICIES |
| Mohamed ALAMI CHEHBOUNE, Ecole Polytechnique / IRT SystemX, France; Rim kaddah, IRT SystemX, France; Luca Martino, Universidad Rey Juan Carlos, Spain; Fernando Llorente, Universidad Carlos III de Madrid, Spain; Jesse Read, Ecole Polytechnique, France |
| SiG-DML_L5.4: MULTI-HEAD TEMPORAL ATTENTION-AUGMENTED BILINEAR NETWORK FOR FINANCIAL TIME SERIES PREDICTION |
| Mostafa Shabani, Martin Magris, Alexandros Iosifidis, Aarhus University, Denmark; Dat Thanh Tran, Juho Kanniainen, Tampere University, Finland |
| SiG-DML_L5.5: TIME-VARYING NORMALIZING FLOW FOR GENERATIVE MODELING OF DYNAMICAL SIGNALS |
| Anubhab Ghosh, Aleix Espuna Fontcuberta, Saikat Chatterjee, KTH Royal Institute of Technology, Sweden; Mohamed R. -H. Abdalmoaty, Uppsala University, Sweden |