Alessandro Salatiello
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Alessandro Salatiello has a total of 48 co-authors. Here are the top co-authors by number of shared publications:
- Martin A. Giese
Section Computational Sensomotorics, Hertie Institute for Clinical Brain Research, 72076 Tuebingen, Germany
3 shared publications - Akilles Rechardt
Department of Psychology, Royal Holloway, University of London, Egham, United Kingdom
1 shared publication - Alexandra O. Cohen
Department of Psychology, Emory University, Atlanta, GA, USA
1 shared publication - Ankush Chakrabarty
Mitsubishi Electric Research Laboratories
1 shared publication - Anna Behler
Hunter Medical Research Institute, New Lambton Heights, NSW, Australia
1 shared publication Anton Pashkov
Department of Neurosurgery, Novosibirsk State Medical University, Novosibirsk, Russia
1 shared publication
Latest Publications
- Open Access
From Modules to Movement: Deconstructing the Modular Architecture of the Motor System
Coordinating multi-articulated bodies to generate purposeful movement is a formidable computational challenge. Yet the human motor system performs this task robustly in dynamic, uncertain environments, despite noisy and delayed feedback, slow actuators, and strict energetic constraints. A central question is what organizational principles underlie this efficiency. One widely recognized principle of neural organization is modularity, which enables complex problems to be decomposed into simpler subproblems that specialized modules are optimized to solve. In this review, we argue that modularity is a fundamental organizing principle of the motor system. We first summarize evidence for brain modularity, ranging from classical lesion studies to contemporary graph-theoretical analyses. We next discuss the main factors underlying the emergence and evolutionary selection of modular architectures, highlighting the computational advantages they provide. We then review the major neuroanatomical modules that structure current descriptions of the motor system and compare three prominent computational frameworks of motor control$-$optimal feedback control theory, muscle synergy theory, and dynamical systems approaches$-$showing that all implicitly or explicitly rely on specialized computational modules. We conclude by contrasting the key strengths and limitations of existing frameworks and by proposing promising directions toward more comprehensive theories.
Open AccessFrom Modules to Movement: Deconstructing the Modular Architecture of the Motor System
Coordinating multi-articulated bodies to generate purposeful movement is a formidable computational challenge. Yet the human motor system performs this task robustly in dynamic, uncertain environments, despite noisy and delayed feedback, slow actuators, and strict energetic constraints. A central question is what organizational principles underlie this efficiency. One widely recognized principle of neural organization is modularity, which enables complex problems to be decomposed into simpler subproblems that specialized modules are optimized to solve. In this review, we argue that modularity is a fundamental organizing principle of the motor system. We first summarize evidence for brain modularity, ranging from classical lesion studies to contemporary graph-theoretical analyses. We next discuss the main factors underlying the emergence and evolutionary selection of modular architectures, highlighting the computational advantages they provide. We then review the major neuroanatomical modules that structure current descriptions of the motor system and compare three prominent computational frameworks of motor control$-$optimal feedback control theory, muscle synergy theory, and dynamical systems approaches$-$showing that all implicitly or explicitly rely on specialized computational modules. We conclude by contrasting the key strengths and limitations of existing frameworks and by proposing promising directions toward more comprehensive theories.
- Open Access
Modularity is the Bedrock of Natural and Artificial Intelligence
The remarkable performance of modern AI systems has been driven by unprecedented scales of data, computation, and energy -- far exceeding the resources required by human intelligence. This disparity highlights the need for new guiding principles and motivates drawing inspiration from the fundamental organizational principles of brain computation. Among these principles, modularity has been shown to be critical for supporting the efficient learning and strong generalization abilities consistently exhibited by humans. Furthermore, modularity aligns well with the No Free Lunch Theorem, which highlights the need for problem-specific inductive biases and motivates architectures composed of specialized components that solve subproblems. However, despite its fundamental role in natural intelligence and its demonstrated benefits across a range of seemingly disparate AI subfields, modularity remains relatively underappreciated in mainstream AI research. In this work, we review several research threads in artificial intelligence and neuroscience through a conceptual framework that highlights the central role of modularity in supporting both artificial and natural intelligence. In particular, we examine what computational advantages modularity provides, how it has emerged as a solution across several AI research areas, which modularity principles the brain exploits, and how modularity can help bridge the gap between natural and artificial intelligence.
Open AccessModularity is the Bedrock of Natural and Artificial Intelligence
The remarkable performance of modern AI systems has been driven by unprecedented scales of data, computation, and energy -- far exceeding the resources required by human intelligence. This disparity highlights the need for new guiding principles and motivates drawing inspiration from the fundamental organizational principles of brain computation. Among these principles, modularity has been shown to be critical for supporting the efficient learning and strong generalization abilities consistently exhibited by humans. Furthermore, modularity aligns well with the No Free Lunch Theorem, which highlights the need for problem-specific inductive biases and motivates architectures composed of specialized components that solve subproblems. However, despite its fundamental role in natural intelligence and its demonstrated benefits across a range of seemingly disparate AI subfields, modularity remains relatively underappreciated in mainstream AI research. In this work, we review several research threads in artificial intelligence and neuroscience through a conceptual framework that highlights the central role of modularity in supporting both artificial and natural intelligence. In particular, we examine what computational advantages modularity provides, how it has emerged as a solution across several AI research areas, which modularity principles the brain exploits, and how modularity can help bridge the gap between natural and artificial intelligence.
- Open Access
Hierarchical Time Series Forecasting Via Latent Mean Encoding
Coherently forecasting the behaviour of a target variable across both coarse and fine temporal scales is crucial for profit-optimized decision-making in several business applications, and remains an open research problem in temporal hierarchical forecasting. Here, we propose a new hierarchical architecture that tackles this problem by leveraging modules that specialize in forecasting the different temporal aggregation levels of interest. The architecture, which learns to encode the average behaviour of the target variable within its hidden layers, makes accurate and coherent forecasts across the target temporal hierarchies. We validate our architecture on the challenging, real-world M5 dataset and show that it outperforms established methods, such as the TSMixer model.
Open AccessHierarchical Time Series Forecasting Via Latent Mean Encoding
Coherently forecasting the behaviour of a target variable across both coarse and fine temporal scales is crucial for profit-optimized decision-making in several business applications, and remains an open research problem in temporal hierarchical forecasting. Here, we propose a new hierarchical architecture that tackles this problem by leveraging modules that specialize in forecasting the different temporal aggregation levels of interest. The architecture, which learns to encode the average behaviour of the target variable within its hidden layers, makes accurate and coherent forecasts across the target temporal hierarchies. We validate our architecture on the challenging, real-world M5 dataset and show that it outperforms established methods, such as the TSMixer model.
- Top Cited PapersOpen Access
- 27 November 2024
- Published by Springer Nature in Nature Human Behaviour
- Vol. 9 (2) , 305-315
Large language models surpass human experts in predicting neuroscience results
Scientific discoveries often hinge on synthesizing decades of research, a task that potentially outstrips human information processing capacities. Large language models (LLMs) offer a solution. LLMs trained on the vast scientific literature could potentially integrate noisy yet interrelated findings to forecast novel results better than human experts. Here, to evaluate this possibility, we created BrainBench, a forward-looking benchmark for predicting neuroscience results. We find that LLMs surpass experts in predicting experimental outcomes. BrainGPT, an LLM we tuned on the neuroscience literature, performed better yet. Like human experts, when LLMs indicated high confidence in their predictions, their responses were more likely to be correct, which presages a future where LLMs assist humans in making discoveries. Our approach is not neuroscience specific and is transferable to other knowledge-intensive endeavours.
- 27 November 2024
- Published by Springer Nature in Nature Human Behaviour
- Vol. 9 (2) , 305-315
Top Cited PapersOpen AccessLarge language models surpass human experts in predicting neuroscience results
Scientific discoveries often hinge on synthesizing decades of research, a task that potentially outstrips human information processing capacities. Large language models (LLMs) offer a solution. LLMs trained on the vast scientific literature could potentially integrate noisy yet interrelated findings to forecast novel results better than human experts. Here, to evaluate this possibility, we created BrainBench, a forward-looking benchmark for predicting neuroscience results. We find that LLMs surpass experts in predicting experimental outcomes. BrainGPT, an LLM we tuned on the neuroscience literature, performed better yet. Like human experts, when LLMs indicated high confidence in their predictions, their responses were more likely to be correct, which presages a future where LLMs assist humans in making discoveries. Our approach is not neuroscience specific and is transferable to other knowledge-intensive endeavours.
- Open Access
- 20 June 2023
- Published by Association for Computing Machinery (ACM)
- p. 125-133
Synthesizing Building Operation Data with Generative Models: VAEs, GANs, or Something In Between?
The generation of time-series profiles of building operation requires expensive and time-consuming data consolidation and modeling efforts that rely on extensive domain knowledge and need frequent revisions due to evolving energy systems, user behavior, and environmental conditions. Generative deep learning may be used to provide an automatic, scalable, data-source-agnostic, and efficient method to synthesize these artificial time-series profiles by learning the distribution of the original data. While a range of generative neural networks have been proposed, generative adversarial networks (GANs) and variational autoencoders (VAEs) are most popular models; GANs typically require considerable customization to stabilize the training procedure, while VAEs are often reported to generate lower-quality samples compared to GANs. In this paper, we propose a network architecture and training procedure that combines the strengths of VAEs and GANs by incorporating Regularized Adversarial Fine-Tuning (RAFT). We imbue the architecture with conditional inputs to reflect ambient/outdoor conditions and operating conditions, and demonstrate its effectiveness by using operational data collected over 585 days from SUSTIE: Mitsubishi Electric’s net-zero energy building. Comparing against classical GAN, VAE, Wasserstein-GAN, and VAE-GAN, our proposed conditional RAFT-VAE-GAN outperforms its competitors in terms of mean accuracy, training stability, and several metrics that ascertain how close the synthetic distribution is to the measured data distribution.
- 20 June 2023
- Published by Association for Computing Machinery (ACM)
- p. 125-133
Open AccessSynthesizing Building Operation Data with Generative Models: VAEs, GANs, or Something In Between?
The generation of time-series profiles of building operation requires expensive and time-consuming data consolidation and modeling efforts that rely on extensive domain knowledge and need frequent revisions due to evolving energy systems, user behavior, and environmental conditions. Generative deep learning may be used to provide an automatic, scalable, data-source-agnostic, and efficient method to synthesize these artificial time-series profiles by learning the distribution of the original data. While a range of generative neural networks have been proposed, generative adversarial networks (GANs) and variational autoencoders (VAEs) are most popular models; GANs typically require considerable customization to stabilize the training procedure, while VAEs are often reported to generate lower-quality samples compared to GANs. In this paper, we propose a network architecture and training procedure that combines the strengths of VAEs and GANs by incorporating Regularized Adversarial Fine-Tuning (RAFT). We imbue the architecture with conditional inputs to reflect ambient/outdoor conditions and operating conditions, and demonstrate its effectiveness by using operational data collected over 585 days from SUSTIE: Mitsubishi Electric’s net-zero energy building. Comparing against classical GAN, VAE, Wasserstein-GAN, and VAE-GAN, our proposed conditional RAFT-VAE-GAN outperforms its competitors in terms of mean accuracy, training stability, and several metrics that ascertain how close the synthetic distribution is to the measured data distribution.
- Open Access
- 18 July 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-8
Continuous Decoding of Daily-Life Hand Movements from Forearm Muscle Activity for Enhanced Myoelectric Control of Hand Prostheses
State-of-the-art motorized hand prostheses are endowed with actuators able to provide independent and proportional control of as many as six degrees of freedom (DOFs). The control signals are derived from residual electromyographic (EMG) activity, recorded concurrently from relevant forearm muscles. Nevertheless, the functional mapping between forearm EMG activity and hand kinematics is only known with limited accuracy. Therefore, no robust method exists for the reliable computation of control signals for the independent and proportional actuation of more than two DOFs. A common approach to deal with this limitation is to preprogram the prostheses for the execution of a restricted number of behaviors (e.g., pinching, grasping, and wrist rotation) that are activated by the detection of specific EMG activation patterns. However, this approach severely limits the range of activities users can perform with the prostheses during their daily living. In this work, we introduce a novel method, based on a long short-term memory (LSTM) network, to map forearm EMG activity onto hand kinematics online. Critically, unlike previous research efforts that tend to focus on simple and highly controlled motor tasks, we tested our method on a dataset of daily living activities (ADLs): the KIN-MUS UJI dataset. To the best of our knowledge, ours is the first reported work on the prediction of hand kinematics that uses this challenging dataset. Remarkably, we show that our network is able to generalize to novel untrained ADLs. Our results suggest that the presented method is suitable for the generation of control signals for the independent and proportional actuation of the multiple DOFs of state-of-the-art hand prostheses.
- 18 July 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-8
Open AccessContinuous Decoding of Daily-Life Hand Movements from Forearm Muscle Activity for Enhanced Myoelectric Control of Hand Prostheses
State-of-the-art motorized hand prostheses are endowed with actuators able to provide independent and proportional control of as many as six degrees of freedom (DOFs). The control signals are derived from residual electromyographic (EMG) activity, recorded concurrently from relevant forearm muscles. Nevertheless, the functional mapping between forearm EMG activity and hand kinematics is only known with limited accuracy. Therefore, no robust method exists for the reliable computation of control signals for the independent and proportional actuation of more than two DOFs. A common approach to deal with this limitation is to preprogram the prostheses for the execution of a restricted number of behaviors (e.g., pinching, grasping, and wrist rotation) that are activated by the detection of specific EMG activation patterns. However, this approach severely limits the range of activities users can perform with the prostheses during their daily living. In this work, we introduce a novel method, based on a long short-term memory (LSTM) network, to map forearm EMG activity onto hand kinematics online. Critically, unlike previous research efforts that tend to focus on simple and highly controlled motor tasks, we tested our method on a dataset of daily living activities (ADLs): the KIN-MUS UJI dataset. To the best of our knowledge, ours is the first reported work on the prediction of hand kinematics that uses this challenging dataset. Remarkably, we show that our network is able to generalize to novel untrained ADLs. Our results suggest that the presented method is suitable for the generation of control signals for the independent and proportional actuation of the multiple DOFs of state-of-the-art hand prostheses.
- Open Access
- 14 October 2020
- Published by Springer Nature
- p. 874-886
Recurrent Neural Network Learning of Performance and Intrinsic Population Dynamics from Sparse Neural Data
- 14 October 2020
- Published by Springer Nature
- p. 874-886
Open AccessRecurrent Neural Network Learning of Performance and Intrinsic Population Dynamics from Sparse Neural Data
- Open Access
Recurrent Neural Network Learning of Performance and Intrinsic Population Dynamics from Sparse Neural Data
Recurrent Neural Networks (RNNs) are popular models of brain function. The typical training strategy is to adjust their input-output behavior so that it matches that of the biological circuit of interest. Even though this strategy ensures that the biological and artificial networks perform the same computational task, it does not guarantee that their internal activity dynamics match. This suggests that the trained RNNs might end up performing the task employing a different internal computational mechanism, which would make them a suboptimal model of the biological circuit. In this work, we introduce a novel training strategy that allows learning not only the input-output behavior of an RNN but also its internal network dynamics, based on sparse neural recordings. We test the proposed method by training an RNN to simultaneously reproduce internal dynamics and output signals of a physiologically-inspired neural model. Specifically, this model generates the multiphasic muscle-like activity patterns typically observed during the execution of reaching movements, based on the oscillatory activation patterns concurrently observed in the motor cortex. Remarkably, we show that the reproduction of the internal dynamics is successful even when the training algorithm relies on the activities of a small subset of neurons sampled from the biological network. Furthermore, we show that training the RNNs with this method significantly improves their generalization performance. Overall, our results suggest that the proposed method is suitable for building powerful functional RNN models, which automatically capture important computational properties of the biological circuit of interest from sparse neural recordings.
Open AccessRecurrent Neural Network Learning of Performance and Intrinsic Population Dynamics from Sparse Neural Data
Recurrent Neural Networks (RNNs) are popular models of brain function. The typical training strategy is to adjust their input-output behavior so that it matches that of the biological circuit of interest. Even though this strategy ensures that the biological and artificial networks perform the same computational task, it does not guarantee that their internal activity dynamics match. This suggests that the trained RNNs might end up performing the task employing a different internal computational mechanism, which would make them a suboptimal model of the biological circuit. In this work, we introduce a novel training strategy that allows learning not only the input-output behavior of an RNN but also its internal network dynamics, based on sparse neural recordings. We test the proposed method by training an RNN to simultaneously reproduce internal dynamics and output signals of a physiologically-inspired neural model. Specifically, this model generates the multiphasic muscle-like activity patterns typically observed during the execution of reaching movements, based on the oscillatory activation patterns concurrently observed in the motor cortex. Remarkably, we show that the reproduction of the internal dynamics is successful even when the training algorithm relies on the activities of a small subset of neurons sampled from the biological network. Furthermore, we show that training the RNNs with this method significantly improves their generalization performance. Overall, our results suggest that the proposed method is suitable for building powerful functional RNN models, which automatically capture important computational properties of the biological circuit of interest from sparse neural recordings.