arXiv subjects

Alberto Silvio Chiappa

Publications and source records attributed to Alberto Silvio Chiappa.

7 recordsLinked to original sources

arXiv2026

Arnold: A multi-task, multi-embodiment muscle transformer policy

Controlling high-dimensional and nonlinear musculoskeletal models of the human body is a foundational scientific challenge. Recent machine learning breakthroughs have heralded in-silico policies that master individual skills like reaching, object manipulation and locomotion in musculoskeletal systems with many degrees of freedom. However, these agents are merely "specialists", achieving high performance for a single skill. In this work, we develop Arnold, a transformer-based musculoskeletal control policy that masters multiple tasks and embodiments. Arnold combines behavior cloning and reinforcement learning to address 14 challenging control tasks spanning dexterous object manipulation, reaching, and locomotion, matching or exceeding the performance of single-task specialist policies. A key innovation is Arnold's sensorimotor vocabulary, a compositional representation of the semantics of heterogeneous sensory modalities, objectives, and actuators. Arnold leverages this vocabulary via a transformer architecture to deal with the variable observation and action spaces across tasks. This framework supports efficient multi-task, multi-embodiment learning and facilitates rapid adaptation to novel tasks, while encouraging universal motor strategies such as action and kinematic smoothness. Finally, causal probing of the motor output reveals that low-dimensional muscle synergies remain largely task-specific and that variance-based analyses systematically underestimate functional control dimensionality, consistent with biological observations on the limited transferability of such synergies. Code and data are available here: https://github.com/amathislab/arnold

cs.RO↗

arXiv2026

KINESIS: Motion Imitation for Human Musculoskeletal Locomotion

How do humans move? Advances in reinforcement learning (RL) have produced impressive results in capturing human motion using physics-based humanoid control. However, torque-controlled humanoids fail to model key aspects of human motor control such as biomechanical joint constraints & non-linear and overactuated musculotendon control. We present KINESIS, a model-free motion imitation framework that tackles these challenges. KINESIS is trained on 1.8 hours of locomotion data and achieves strong motion imitation performance on unseen trajectories. Through a negative mining approach, KINESIS learns robust locomotion priors that we leverage to deploy the policy on several downstream tasks such as text-to-control, target point reaching, and football penalty kicks. Importantly, KINESIS learns to generate muscle activity patterns that correlate well with human EMG activity. We show that these results scale seamlessly across biomechanical model complexity, demonstrating control of up to 290 muscles. Overall, the physiological plausibility makes KINESIS a promising model for tackling challenging problems in human motor control. Code, videos and benchmarks are available at https://github.com/amathislab/Kinesis.

cs.RO↗

arXiv2025

Auto-bidding in real-time auctions via Oracle Imitation Learning (OIL)

Online advertising has become one of the most successful business models of the internet era. Impression opportunities are typically allocated through real-time auctions, where advertisers bid to secure advertisement slots. Deciding the best bid for an impression opportunity is challenging, due to the stochastic nature of user behavior and the variability of advertisement traffic over time. In this work, we propose a framework for training auto-bidding agents in multi-slot second-price auctions to maximize acquisitions (e.g., clicks, conversions) while adhering to budget and cost-per-acquisition (CPA) constraints. We exploit the insight that, after an advertisement campaign concludes, determining the optimal bids for each impression opportunity can be framed as a multiple-choice knapsack problem (MCKP) with a nonlinear objective. We propose an "oracle" algorithm that identifies a near-optimal combination of impression opportunities and advertisement slots, considering both past and future advertisement traffic data. This oracle solution serves as a training target for a student network which bids having access only to real-time information, a method we term Oracle Imitation Learning (OIL). Through numerical experiments, we demonstrate that OIL achieves superior performance compared to both online and offline reinforcement learning algorithms, offering improved sample efficiency. Notably, OIL shifts the complexity of training auto-bidding agents from crafting sophisticated learning algorithms to solving a nonlinear constrained optimization problem efficiently.

cs.LG↗

arXiv2025

Adaptive Budget Optimization for Multichannel Advertising Using Combinatorial Bandits

Effective budget allocation is crucial for optimizing the performance of digital advertising campaigns. However, the development of practical budget allocation algorithms remain limited, primarily due to the lack of public datasets and comprehensive simulation environments capable of verifying the intricacies of real-world advertising. While multi-armed bandit (MAB) algorithms have been extensively studied, their efficacy diminishes in non-stationary environments where quick adaptation to changing market dynamics is essential. In this paper, we advance the field of budget allocation in digital advertising by introducing three key contributions. First, we develop a simulation environment designed to mimic multichannel advertising campaigns over extended time horizons, incorporating logged real-world data. Second, we propose an enhanced combinatorial bandit budget allocation strategy that leverages a saturating mean function and a targeted exploration mechanism with change-point detection. This approach dynamically adapts to changing market conditions, improving allocation efficiency by filtering target regions based on domain knowledge. Finally, we present both theoretical analysis and empirical results, demonstrating that our method consistently outperforms baseline strategies, achieving higher rewards and lower regret across multiple real-world campaigns.

cs.LG↗

arXiv2023

Latent Exploration for Reinforcement Learning

In Reinforcement Learning, agents learn policies by exploring and interacting with the environment. Due to the curse of dimensionality, learning policies that map high-dimensional sensory input to motor output is particularly challenging. During training, state of the art methods (SAC, PPO, etc.) explore the environment by perturbing the actuation with independent Gaussian noise. While this unstructured exploration has proven successful in numerous tasks, it can be suboptimal for overactuated systems. When multiple actuators, such as motors or muscles, drive behavior, uncorrelated perturbations risk diminishing each other's effect, or modifying the behavior in a task-irrelevant way. While solutions to introduce time correlation across action perturbations exist, introducing correlation across actuators has been largely ignored. Here, we propose LATent TIme-Correlated Exploration (Lattice), a method to inject temporally-correlated noise into the latent state of the policy network, which can be seamlessly integrated with on- and off-policy algorithms. We demonstrate that the noisy actions generated by perturbing the network's activations can be modeled as a multivariate Gaussian distribution with a full covariance matrix. In the PyBullet locomotion tasks, Lattice-SAC achieves state of the art results, and reaches 18% higher reward than unstructured exploration in the Humanoid environment. In the musculoskeletal control environments of MyoSuite, Lattice-PPO achieves higher reward in most reaching and object manipulation tasks, while also finding more energy-efficient policies with reductions of 20-60%. Overall, we demonstrate the effectiveness of structured action noise in time and actuator space for complex motor control tasks. The code is available at: https://github.com/amathislab/lattice.

cs.RO↗

arXiv2023

Reducing the Drag of a Bluff Body by Deep Reinforcement Learning

We present a deep reinforcement learning approach to a classical problem in fluid dynamics, i.e., the reduction of the drag of a bluff body. We cast the problem as a discrete-time control with continuous action space: at each time step, an autonomous agent can set the flow rate of two jets of fluid, positioned at the back of the body. The agent, trained with Proximal Policy Optimization, learns an effective strategy to make the jets interact with the vortexes of the wake, thus reducing the drag. To tackle the computational complexity of the fluid dynamics simulations, which would make the training procedure prohibitively expensive, we train the agent on a coarse discretization of the domain. We provide numerical evidence that a policy trained in this approximate environment still retains good performance when carried over to a denser mesh. Our simulations show a considerable drag reduction with a consequent saving of total power, defined as the sum of the power spent by the control system and of the power of the drag force, amounting to 40% when compared to simulations with the reference bluff body without any jet. Finally, we qualitatively investigate the control policy learnt by the neural network. We can observe that it achieves the drag reduction by learning the frequency of formation of the vortexes and activating the jets accordingly, thus blowing them away off the rear body surface.

physics.flu-dyn↗

arXiv2022

DMAP: a Distributed Morphological Attention Policy for Learning to Locomote with a Changing Body

Biological and artificial agents need to deal with constant changes in the real world. We study this problem in four classical continuous control environments, augmented with morphological perturbations. Learning to locomote when the length and the thickness of different body parts vary is challenging, as the control policy is required to adapt to the morphology to successfully balance and advance the agent. We show that a control policy based on the proprioceptive state performs poorly with highly variable body configurations, while an (oracle) agent with access to a learned encoding of the perturbation performs significantly better. We introduce DMAP, a biologically-inspired, attention-based policy network architecture. DMAP combines independent proprioceptive processing, a distributed policy with individual controllers for each joint, and an attention mechanism, to dynamically gate sensory information from different body parts to different controllers. Despite not having access to the (hidden) morphology information, DMAP can be trained end-to-end in all the considered environments, overall matching or surpassing the performance of an oracle agent. Thus DMAP, implementing principles from biological motor control, provides a strong inductive bias for learning challenging sensorimotor tasks. Overall, our work corroborates the power of these principles in challenging locomotion tasks.

cs.RO↗