{"product_id":"interpretable-representation-learning-for-motion-forecasting-von-royden-wagner","title":"Interpretable Representation Learning for Motion Forecasting","description":"\n                                \n                \u003cp\u003eWe address interpretable representation learning for motion forecasting in self-driving cars. Rather than treating transformers as black boxes, we develop methods to interpret and modify learned representations. We introduce self-supervised pre-training with interpretable objectives. Moreover, we probe latent spaces of forecasting models and reveal interpretable features, allowing us to make targeted interventions. Finally, we uncover retrocausal mechanisms, which enable goal-based instructions.\u003c\/p\u003e\n                            \n            \u003cdiv class=\"aw-variant-hidden-subtitle-div\" id=\"aw-variant-subtitle-9783731514749\"\u003e\u003ch3\u003e\u003c\/h3\u003e\u003c\/div\u003e","brand":"Autorenwelt Shop","offers":[{"title":"Softcover - 9783731514749","offer_id":57978637517125,"sku":"9783731514749","price":38.0,"currency_code":"EUR","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0940\/0622\/files\/163e8dd3-4e17-47e6-be16-9e2aa50c32e3.jpg?v=1779341097","url":"https:\/\/shop.autorenwelt.de\/en\/products\/interpretable-representation-learning-for-motion-forecasting-von-royden-wagner","provider":"Autorenwelt Shop","version":"1.0","type":"link"}