Arunkumar Byravan
YOU?
Author Swipe
View article: Gemini Robotics 1.5: Pushing the Frontier of Generalist Robots with Advanced Embodied Reasoning, Thinking, and Motion Transfer
Gemini Robotics 1.5: Pushing the Frontier of Generalist Robots with Advanced Embodied Reasoning, Thinking, and Motion Transfer Open
General-purpose robots need a deep understanding of the physical world, advanced reasoning, and general and dexterous control. This report introduces the latest generation of the Gemini Robotics model family: Gemini Robotics 1.5, a multi-e…
View article: Splatting Physical Scenes: End-to-End Real-to-Sim from Imperfect Robot Data
Splatting Physical Scenes: End-to-End Real-to-Sim from Imperfect Robot Data Open
Creating accurate, physical simulations directly from real-world robot motion holds great value for safe, scalable, and affordable robot learning, yet remains exceptionally challenging. Real robot data suffers from occlusions, noisy camera…
View article: Gemini Robotics: Bringing AI into the Physical World
Gemini Robotics: Bringing AI into the Physical World Open
Recent advancements in large multimodal models have led to the emergence of remarkable generalist capabilities in digital domains, yet their translation to physical agents such as robots remains a significant challenge. This report introdu…
View article: Proc4Gem: Foundation models for physical agency through procedural generation
Proc4Gem: Foundation models for physical agency through procedural generation Open
In robot learning, it is common to either ignore the environment semantics, focusing on tasks like whole-body control which only require reasoning about robot-environment contacts, or conversely to ignore contact dynamics, focusing on grou…
View article: Diffusion Augmented Agents: A Framework for Efficient Exploration and Transfer Learning
Diffusion Augmented Agents: A Framework for Efficient Exploration and Transfer Learning Open
We introduce Diffusion Augmented Agents (DAAG), a novel framework that leverages large language models, vision language models, and diffusion models to improve sample efficiency and transfer learning in reinforcement learning for embodied …
View article: Learning Robot Soccer from Egocentric Vision with Deep Reinforcement Learning
Learning Robot Soccer from Egocentric Vision with Deep Reinforcement Learning Open
We apply multi-agent deep reinforcement learning (RL) to train end-to-end robot soccer policies with fully onboard computation and sensing via egocentric RGB vision. This setting reflects many challenges of real-world robotics, including a…
View article: Real-World Fluid Directed Rigid Body Control via Deep Reinforcement Learning
Real-World Fluid Directed Rigid Body Control via Deep Reinforcement Learning Open
Recent advances in real-world applications of reinforcement learning (RL) have relied on the ability to accurately simulate systems at scale. However, domains such as fluid dynamical systems exhibit complex dynamic phenomena that are hard …
View article: Foundations for Transfer in Reinforcement Learning: A Taxonomy of Knowledge Modalities
Foundations for Transfer in Reinforcement Learning: A Taxonomy of Knowledge Modalities Open
Contemporary artificial intelligence systems exhibit rapidly growing abilities accompanied by the growth of required resources, expansive datasets and corresponding investments into computing infrastructure. Although earlier successes pred…
View article: Equivariant Data Augmentation for Generalization in Offline Reinforcement Learning
Equivariant Data Augmentation for Generalization in Offline Reinforcement Learning Open
We present a novel approach to address the challenge of generalization in offline reinforcement learning (RL), where the agent learns from a fixed dataset without any additional interaction with the environment. Specifically, we aim to imp…
View article: Towards A Unified Agent with Foundation Models
Towards A Unified Agent with Foundation Models Open
Language Models and Vision Language Models have recently demonstrated unprecedented capabilities in terms of understanding human intentions, reasoning, scene understanding, and planning-like behaviour, in text form, among many others. In t…
View article: A Generalist Dynamics Model for Control
A Generalist Dynamics Model for Control Open
We investigate the use of transformer sequence models as dynamics models (TDMs) for control. We find that TDMs exhibit strong generalization capabilities to unseen environments, both in a few-shot setting, where a generalist TDM is fine-tu…
View article: Learning Agile Soccer Skills for a Bipedal Robot with Deep Reinforcement Learning
Learning Agile Soccer Skills for a Bipedal Robot with Deep Reinforcement Learning Open
We investigate whether Deep Reinforcement Learning (Deep RL) is able to synthesize sophisticated and safe movement skills for a low-cost, miniature humanoid robot that can be composed into complex behavioral strategies in dynamic environme…
View article: Leveraging Jumpy Models for Planning and Fast Learning in Robotic Domains
Leveraging Jumpy Models for Planning and Fast Learning in Robotic Domains Open
In this paper we study the problem of learning multi-step dynamics prediction models (jumpy models) from unlabeled experience and their utility for fast inference of (high-level) plans in downstream tasks. In particular we propose to learn…
View article: NeRF2Real: Sim2real Transfer of Vision-guided Bipedal Motion Skills using Neural Radiance Fields
NeRF2Real: Sim2real Transfer of Vision-guided Bipedal Motion Skills using Neural Radiance Fields Open
We present a system for applying sim2real approaches to "in the wild" scenes with realistic visuals, and to policies which rely on active perception using RGB cameras. Given a short video of a static scene collected using a generic phone, …
View article: Revisiting Gaussian mixture critics in off-policy reinforcement learning: a sample-based approach
Revisiting Gaussian mixture critics in off-policy reinforcement learning: a sample-based approach Open
Actor-critic algorithms that make use of distributional policy evaluation have frequently been shown to outperform their non-distributional counterparts on many challenging control tasks. Examples of this behavior include the D4PG and DMPO…
View article: The Challenges of Exploration for Offline Reinforcement Learning
The Challenges of Exploration for Offline Reinforcement Learning Open
Offline Reinforcement Learning (ORL) enablesus to separately study the two interlinked processes of reinforcement learning: collecting informative experience and inferring optimal behaviour. The second step has been widely studied in the o…
View article: Beyond Pick-and-Place: Tackling Robotic Stacking of Diverse Shapes
Beyond Pick-and-Place: Tackling Robotic Stacking of Diverse Shapes Open
We study the problem of robotic stacking with objects of complex geometry. We propose a challenging and diverse set of such objects that was carefully designed to require strategies beyond a simple "pick-and-place" solution. Our method is …
View article: Evaluating model-based planning and planner amortization for continuous control
Evaluating model-based planning and planner amortization for continuous control Open
There is a widespread intuition that model-based control methods should be able to surpass the data efficiency of model-free approaches. In this paper we attempt to evaluate this intuition on various challenging locomotion tasks. We take a…
View article: Learning Dynamics Models for Model Predictive Agents
Learning Dynamics Models for Model Predictive Agents Open
Model-Based Reinforcement Learning involves learning a \textit{dynamics model} from data, and then using this model to optimise behaviour, most often with an online \textit{planner}. Much of the recent research along these lines presents a…
View article: On Multi-objective Policy Optimization as a Tool for Reinforcement Learning.
On Multi-objective Policy Optimization as a Tool for Reinforcement Learning. Open
Many advances that have improved the robustness and efficiency of deep reinforcement learning (RL) algorithms can, in one way or another, be understood as introducing additional objectives, or constraints, in the policy optimization step. …
View article: On Multi-objective Policy Optimization as a Tool for Reinforcement Learning: Case Studies in Offline RL and Finetuning
On Multi-objective Policy Optimization as a Tool for Reinforcement Learning: Case Studies in Offline RL and Finetuning Open
Many advances that have improved the robustness and efficiency of deep reinforcement learning (RL) algorithms can, in one way or another, be understood as introducing additional objectives or constraints in the policy optimization step. Th…
View article: Representation Matters: Improving Perception and Exploration for Robotics
Representation Matters: Improving Perception and Exploration for Robotics Open
Projecting high-dimensional environment observations into lower-dimensional structured representations can considerably improve data-efficiency for reinforcement learning in domains with limited data such as robotics. Can a single generall…
View article: Representation Matters: Improving Perception and Exploration for\n Robotics
Representation Matters: Improving Perception and Exploration for\n Robotics Open
Projecting high-dimensional environment observations into lower-dimensional\nstructured representations can considerably improve data-efficiency for\nreinforcement learning in domains with limited data such as robotics. Can a\nsingle gener…
View article: Local Search for Policy Iteration in Continuous Control
Local Search for Policy Iteration in Continuous Control Open
We present an algorithm for local, regularized, policy improvement in reinforcement learning (RL) that allows us to formulate model-based and model-free variants in a single framework. Our algorithm can be interpreted as a natural extensio…
View article: Motion-Nets: 6D Tracking of Unknown Objects in Unseen Environments using RGB
Motion-Nets: 6D Tracking of Unknown Objects in Unseen Environments using RGB Open
In this work, we bridge the gap between recent pose estimation and tracking work to develop a powerful method for robots to track objects in their surroundings. Motion-Nets use a segmentation model to segment the scene, and separate transl…
View article: Imagined Value Gradients: Model-Based Policy Optimization with\n Transferable Latent Dynamics Models
Imagined Value Gradients: Model-Based Policy Optimization with\n Transferable Latent Dynamics Models Open
Humans are masters at quickly learning many complex tasks, relying on an\napproximate understanding of the dynamics of their environments. In much the\nsame way, we would like our learning agents to quickly adapt to new tasks. In\nthis pap…
View article: Imagined Value Gradients: Model-Based Policy Optimization with Transferable Latent Dynamics Models
Imagined Value Gradients: Model-Based Policy Optimization with Transferable Latent Dynamics Models Open
Humans are masters at quickly learning many complex tasks, relying on an approximate understanding of the dynamics of their environments. In much the same way, we would like our learning agents to quickly adapt to new tasks. In this paper,…
View article: Prospection: Interpretable plans from language by predicting the future
Prospection: Interpretable plans from language by predicting the future Open
High-level human instructions often correspond to behaviors with multiple implicit steps. In order for robots to be useful in the real world, they must be able to to reason over both motions and intermediate goals implied by human instruct…
View article: Structured Deep Visual Dynamics Models for Robot Manipulation
Structured Deep Visual Dynamics Models for Robot Manipulation Open
View article: SE3-Pose-Nets: Structured Deep Dynamics Models for Visuomotor Planning and Control
SE3-Pose-Nets: Structured Deep Dynamics Models for Visuomotor Planning and Control Open
In this work, we present an approach to deep visuomotor control using structured deep dynamics models. Our deep dynamics model, a variant of SE3-Nets, learns a low-dimensional pose embedding for visuomotor control via an encoder-decoder st…