Franziska Meier
YOU?
Author Swipe
View article: WorldPlanner: Monte Carlo Tree Search and MPC with Action-Conditioned Visual World Models
WorldPlanner: Monte Carlo Tree Search and MPC with Action-Conditioned Visual World Models Open
Robots must understand their environment from raw sensory inputs and reason about the consequences of their actions in it to solve complex tasks. Behavior Cloning (BC) leverages task-specific human demonstrations to learn this knowledge as…
View article: Fast3R: Towards 3D Reconstruction of 1000+ Images in One Forward Pass
Fast3R: Towards 3D Reconstruction of 1000+ Images in One Forward Pass Open
Multi-view 3D reconstruction remains a core challenge in computer vision, particularly in applications requiring accurate and scalable representations across diverse perspectives. Current leading methods such as DUSt3R employ a fundamental…
View article: What do we learn from a large-scale study of pre-trained visual representations in sim and real environments?
What do we learn from a large-scale study of pre-trained visual representations in sim and real environments? Open
We present a large empirical investigation on the use of pre-trained visual representations (PVRs) for training downstream policies that execute real-world tasks. Our study involves five different PVRs, each trained for five distinct manip…
View article: EgoAdapt: A multi-stream evaluation study of adaptation to real-world egocentric user video
EgoAdapt: A multi-stream evaluation study of adaptation to real-world egocentric user video Open
In egocentric action recognition a single population model is typically trained and subsequently embodied on a head-mounted device, such as an augmented reality headset. While this model remains static for new users and environments, we in…
View article: Where are we in the search for an Artificial Visual Cortex for Embodied Intelligence?
Where are we in the search for an Artificial Visual Cortex for Embodied Intelligence? Open
We present the largest and most comprehensive empirical study of pre-trained visual representations (PVRs) or visual 'foundation models' for Embodied AI. First, we curate CortexBench, consisting of 17 different tasks spanning locomotion, n…
View article: BC-IRL: Learning Generalizable Reward Functions from Demonstrations
BC-IRL: Learning Generalizable Reward Functions from Demonstrations Open
How well do reward functions learned with inverse reinforcement learning (IRL) generalize? We illustrate that state-of-the-art IRL algorithms, which maximize a maximum-entropy objective, learn rewards that overfit to the demonstrations. Su…
View article: Neural Constraint Satisfaction: Hierarchical Abstraction for Combinatorial Generalization in Object Rearrangement
Neural Constraint Satisfaction: Hierarchical Abstraction for Combinatorial Generalization in Object Rearrangement Open
Object rearrangement is a challenge for embodied agents because solving these tasks requires generalizing across a combinatorially large set of configurations of entities and their locations. Worse, the representations of these entities ar…
View article: Cross-Domain Transfer via Semantic Skill Imitation
Cross-Domain Transfer via Semantic Skill Imitation Open
We propose an approach for semantic imitation, which uses demonstrations from a source domain, e.g. human videos, to accelerate reinforcement learning (RL) in a different target domain, e.g. a robotic manipulator in a simulated kitchen. In…
View article: Neural Grasp Distance Fields for Robot Manipulation
Neural Grasp Distance Fields for Robot Manipulation Open
We formulate grasp learning as a neural field and present Neural Grasp Distance Fields (NGDF). Here, the input is a 6D pose of a robot end effector and output is a distance to a continuous manifold of valid grasps for an object. In contras…
View article: Supervised learning and reinforcement learning of feedback models for reactive behaviors: Tactile feedback testbed
Supervised learning and reinforcement learning of feedback models for reactive behaviors: Tactile feedback testbed Open
Robots need to be able to adapt to unexpected changes in the environment such that they can autonomously succeed in their tasks. However, hand-designing feedback models for adaptation is tedious, if at all possible, making data-driven meth…
View article: Giorgio Pasquali und die <i>Filologia dantesca</i>
Giorgio Pasquali und die <i>Filologia dantesca</i> Open
Riassunto L’articolo parte dall’ipotesi che l’università di Gottinga, innanzitutto il suo dipartimento di filologia greca e antica, abbia avuto un qualche ruolo nel rinnovamento della filologia dantesca quale Michele Barbi l’aveva richiest…
View article: Model Based Meta Learning of Critics for Policy Gradients
Model Based Meta Learning of Critics for Policy Gradients Open
Being able to seamlessly generalize across different tasks is fundamental for robots to act in our world. However, learning representations that generalize quickly to new scenarios is still an open research problem in reinforcement learnin…
View article: Differentiable and Learnable Robot Models
Differentiable and Learnable Robot Models Open
Building differentiable simulations of physical processes has recently received an increasing amount of attention. Specifically, some efforts develop differentiable robotic physics engines motivated by the computational benefits of merging…
View article: Residual Learning From Demonstration: Adapting DMPs for Contact-Rich Manipulation
Residual Learning From Demonstration: Adapting DMPs for Contact-Rich Manipulation Open
Manipulation skills involving contact and friction are inherent to many robotics tasks. Using the class of motor primitives for peg-in-hole like insertions, we study how robots can learn such skills. Dynamic Movement Primitives (DMP) are a…
View article: Block Contextual MDPs for Continual Learning
Block Contextual MDPs for Continual Learning Open
In reinforcement learning (RL), when defining a Markov Decision Process (MDP), the environment dynamics is implicitly assumed to be stationary. This assumption of stationarity, while simplifying, can be unrealistic in many scenarios. In th…
View article: Wie ich auf Dante gekommen bin
Wie ich auf Dante gekommen bin Open
View article: «Una spada lucida e aguta». La paura di Dante e l’'auctoritas' letteraria di San Paolo nella 'Divina Commedia'
«Una spada lucida e aguta». La paura di Dante e l’'auctoritas' letteraria di San Paolo nella 'Divina Commedia' Open
L’articolo torna a interrogare il ruolo di San Paolo nella Divina Commedia. Diversamente da ricercatori recenti secondo i quali San Paolo serve da modello a Dante pellegrino a cui viene accordata una visione divina, questo articolo indaga …
View article: Huntingtin fibrils with different toxicity, structure, and seeding potential can be interconverted
Huntingtin fibrils with different toxicity, structure, and seeding potential can be interconverted Open
View article: Learning Time-Invariant Reward Functions through Model-Based Inverse\n Reinforcement Learning
Learning Time-Invariant Reward Functions through Model-Based Inverse\n Reinforcement Learning Open
Inverse reinforcement learning is a paradigm motivated by the goal of\nlearning general reward functions from demonstrated behaviours. Yet the notion\nof generality for learnt costs is often evaluated in terms of robustness to\nvarious spa…
View article: Learning Time-Invariant Reward Functions through Model-Based Inverse Reinforcement Learning
Learning Time-Invariant Reward Functions through Model-Based Inverse Reinforcement Learning Open
Inverse reinforcement learning is a paradigm motivated by the goal of learning general reward functions from demonstrated behaviours. Yet the notion of generality for learnt costs is often evaluated in terms of robustness to various spatia…
View article: Habitat 2.0: Training Home Assistants to Rearrange their Habitat
Habitat 2.0: Training Home Assistants to Rearrange their Habitat Open
We introduce Habitat 2.0 (H2.0), a simulation platform for training virtual robots in interactive 3D environments and complex physics-enabled scenarios. We make comprehensive contributions to all levels of the embodied AI stack - data, sim…
View article: Leveraging Forward Model Prediction Error for Learning Control
Leveraging Forward Model Prediction Error for Learning Control Open
Learning for model based control can be sample-efficient and generalize well, however successfully learning models and controllers that represent the problem at hand can be challenging for complex tasks. Using inaccurate models for learnin…
View article: Meta Learning via Learned Loss
Meta Learning via Learned Loss Open
Typically, loss functions, regularization mechanisms and other important aspects of training parametric models are chosen heuristically from a limited set of options. In this paper, we take the first step towards automating this process, w…
View article: Convivio
Convivio Open
View article: Titelei/Inhaltsverzeichnis
Titelei/Inhaltsverzeichnis Open
View article: On the use of simulation in robotics: Opportunities, challenges, and suggestions for moving forward
On the use of simulation in robotics: Opportunities, challenges, and suggestions for moving forward Open
The last five years marked a surge in interest for and use of smart robots, which operate in dynamic and unstructured environments and might interact with humans. We posit that well-validated computer simulation can provide a virtual provi…
View article: Learning Navigation Skills for Legged Robots with Learned Robot Embeddings
Learning Navigation Skills for Legged Robots with Learned Robot Embeddings Open
Recent work has shown results on learning navigation policies for idealized cylinder agents in simulation and transferring them to real wheeled robots. Deploying such navigation policies on legged robots can be challenging due to their com…
View article: Multi-Modal Learning of Keypoint Predictive Models for Visual Object Manipulation
Multi-Modal Learning of Keypoint Predictive Models for Visual Object Manipulation Open
Humans have impressive generalization capabilities when it comes to manipulating objects and tools in completely novel environments. These capabilities are, at least partially, a result of humans having internal models of their bodies and …
View article: Learning Extended Body Schemas from Visual Keypoints for Object Manipulation.
Learning Extended Body Schemas from Visual Keypoints for Object Manipulation. Open
Humans have impressive generalization capabilities when it comes to manipulating objects and tools in completely novel environments. These capabilities are, at least partially, a result of humans having internal models of their bodies and …
View article: Exploring Zero-Shot Emergent Communication in Embodied Multi-Agent Populations
Exploring Zero-Shot Emergent Communication in Embodied Multi-Agent Populations Open
Effective communication is an important skill for enabling information exchange and cooperation in multi-agent settings. Indeed, emergent communication is now a vibrant field of research, with common settings involving discrete cheap-talk …