← Back

Machine Learning

Topic spotlight
TopicNeuro

machine learning

Discover seminars, jobs, and research tagged with machine learning across Neuro.
60 curated items50 Seminars10 Positions
Updated 4 days ago
60 items · machine learning

Latest

60 results
PositionNeuroscience

Prof. Ross Williamson

University of Pittsburgh
Pittsburgh, PA, USA
Dec 21, 2025

The Williamson Laboratory investigates the organization and function of auditory cortical projection systems in behaving mice. We use a variety of state-of-the-art tools to probe the neural circuits of awake mice – these include two-photon calcium imaging and high-channel count electrophysiology (both with single-cell optogenetic perturbations), head-fixed behaviors (including virtual reality), and statistical approaches for neural characterization. Details on the research focus and approaches of the laboratory can be found here: https://www.williamsonlaboratory.com/research/

PositionNeuroscience

Joaquin

Gatsby Unit, the Sainsbury Wellcome Centre for Neural Circuits and Behaviour (SWC) and NeuroGEARS Ltd
London, UK
Dec 21, 2025

We invite applications for a Research Software Engineer (RSE) position with expertise in software development, machine learning, neural data analysis, and experimental control (ideally with the Bonsai ecosystem) to contribute to the recently funded project 'Machine Learning for Neuroscience Experimental Control'. You will contribute to the neuroscience community with advanced machine learning software for experimental control. You will be embedded in the unparalleled research environment of the Gatsby Unit, the SWC and NeuroGEARS, with opportunities to connect with top researchers and engineers.

PositionNeuroscience

Prof. Dr. Tobias Rose

University Hospital Bonn, Institute of Experimental Epileptology and Cognition Research
Bonn, Nordrhein-Westfalen, Germany
Dec 21, 2025

The selected candidate will investigate the 'Encoding of Landmark Stability and Stability of Landmark Encoding'. You will study visual landmark encoding at the intersection of hippocampal, thalamic, and cortical inputs to retrosplenial cortex. You will use cutting-edge miniature two-photon Ca2+ imaging, enabling you to longitudinally record activity in defined, large neuronal populations and long-range afferents in freely moving animals. You will carry out rigorous neuronal and behavioral analyses within the confines of automatized closed-loop tasks tailored for visual navigation. This will involve the application of advanced tools for dense behavioral quantification, including multi-angle videography, inertial motion sensing, and egocentric recording with head-mounted cameras for the reconstruction of retinal input. Our aim is to gain a comprehensive understanding of the immediate and sustained multi-area neuronal representation of visual landmarks during unrestricted behavior. We aim to elucidate the mechanisms through which stable visual landmarks are encoded and the processes by which these representations are stabilized to facilitate robust allocentric navigation.

PositionNeuroscience

Arun Antony MD

Jersey Shore University Medical Center
Jersey Shore University Medical Center, Neptune, New Jersey, USA 07753
Dec 21, 2025

The Neuroscience Institute at Jersey Shore University Medical Center, New Jersey, USA is seeking a postdoctoral fellow to work on basic, clinical, and translational projects in the fields of seizures, epilepsy, human intracranial EEG, signal processing, cognition and consciousness. The fellow will join a multidisciplinary team of five epileptologists, neurosurgeons, epilepsy nurses, nurse practitioners, neuropsychologists and researchers providing holistic care to patients with epilepsy. The postdoctoral fellows will have access to the large clinical, imaging, and EEG databases, and outcome measures of cutting edge treatment modalities within the system for research purposes. The successful candidate will be well versed in data collection, processing, programming and will lead an independent research project working closely with collaborators and publish high-quality research.

PositionNeuroscience

Jörn Diedrichsen

Diedrichsen Lab, Western University
Western University, Canada
Dec 21, 2025

We are looking to recruit a new postdoctoral associate for a large collaborative project on the anatomical development of the human cerebellum. The overall goal of the project is to develop a high-resolution normative model of human cerebellar development across the entire life span. The successful candidate will join the Diedrichsen Lab (Western University, Canada) and will work with a team of colleagues at Erasmus Medical Center, the Donders Institute (Netherlands), McGill, Dalhousie, Sick Kids, and UBC (Canada).

PositionNeuroscience

Matthias H Hennig

The University of Edinburgh
Edinburgh
Dec 21, 2025

We are looking for a postdoctoral researcher to develop new machine learning approaches for the analysis of large-scale extracellular recordings. The position is part of a wider effort to enable new discoveries with state-of-the-art electrode arrays and recording devices, and jointly supervised by Matthias Hennig and Matt Nolan. It offers a great opportunity to work with theoretical and experimental neuroscientists innovating open source tools and software for systems neuroscience.

PositionNeuroscience

N/A

Center for Neuroscience and Cell Biology of the University of Coimbra (CNC-UC)
Coimbra and Cantanhede, University of Coimbra
Dec 21, 2025

The PostDoctoral researcher will conduct research activities in modelling and simulation of reward-modulated prosocial behavior and decision-making. The position is part of a larger effort to uncover the computational and mechanistic bases of prosociality and empathy at the behavioral and circuit levels. The role involves working at the interface between experimental data (animal behavior and electrophysiology) and theoretical modelling, with an emphasis on Multi-Agent Reinforcement Learning and neural population dynamics.

PositionNeuroscience

N/A

University of Chicago
Chicago
Dec 21, 2025

The Grossman Center for Quantitative Biology and Human Behavior at the University of Chicago seeks outstanding applicants for multiple postdoctoral positions in computational and theoretical neuroscience. We especially welcome applicants who develop mathematical approaches, computational models, and machine learning methods to study the brain at the circuits, systems, or cognitive levels. The current faculty members of the Grossman Center to work with are: Brent Doiron’s lab investigates how the cellular and synaptic circuitry of neuronal circuits supports the complex dynamics and computations that are routinely observed in the brain. Jorge Jaramillo’s lab investigates how subcortical structures interact with cortical circuits to subserve cognitive processes such as memory, attention, and decision making. Ramon Nogueira’s lab investigates the geometry of representations as the computational support of cognitive processes like abstraction in noisy artificial and biological neural networks. Marcella Noorman’s lab investigates how properties of synapses, neurons, and circuits shape the neural dynamics that enable flexible and efficient computation. Samuel Muscinelli’s lab studies how the anatomy of brain circuits both governs learning and adapts to it. We combine analytical theory, machine learning, and data analysis, in close collaboration with experimentalists. Appointees will have access to state-of-the-art facilities and multiple opportunities for collaboration with exceptional experimental labs within the Neuroscience Institute, as well as other labs from the departments of Physics, Computer Sciences, and Statistics. The Grossman Center offers competitive postdoctoral salaries in the vibrant and international city of Chicago, and a rich intellectual environment that includes the Argonne National Laboratory and UChicago’s Data Science Institute. The Neuroscience Institute is currently engaged in a major expansion that includes the incorporation of several new faculty members in the next few years.

PositionNeuroscience

Professor Geoffrey J Goodhill

Washington University School of Medicine
St Louis, MO
Dec 21, 2025

The Department of Neuroscience at Washington University School of Medicine is seeking a tenure-track investigator at the level of Assistant Professor to develop an innovative research program in Theoretical/Computational Neuroscience. The successful candidate will join a thriving theoretical/computational neuroscience community at Washington University, including the new Center for Theoretical and Computational Neuroscience. In addition, the Department also has world-class research strengths in systems, circuits and behavior, cellular and molecular neuroscience using a variety of animal models including worms, flies, zebrafish, rodents and non-human primates. The Department’s focus on fundamental neuroscience, outstanding research support facilities, and the depth, breadth and collegiality of our culture provide an exceptional environment to launch your independent research program.

SeminarNeuroscience

Computational Mechanisms of Predictive Processing in Brains and Machines

Dr. Antonino Greco
Hertie Institute for Clinical Brain Research, Germany
Dec 10, 2025

Predictive processing offers a unifying view of neural computation, proposing that brains continuously anticipate sensory input and update internal models based on prediction errors. In this talk, I will present converging evidence for the computational mechanisms underlying this framework across human neuroscience and deep neural networks. I will begin with recent work showing that large-scale distributed prediction-error encoding in the human brain directly predicts how sensory representations reorganize through predictive learning. I will then turn to PredNet, a popular predictive coding inspired deep network that has been widely used to model real-world biological vision systems. Using dynamic stimuli generated with our Spatiotemporal Style Transfer algorithm, we demonstrate that PredNet relies primarily on low-level spatiotemporal structure and remains insensitive to high-level content, revealing limits in its generalization capacity. Finally, I will discuss new recurrent vision models that integrate top-down feedback connections with intrinsic neural variability, uncovering a dual mechanism for robust sensory coding in which neural variability decorrelates unit responses, while top-down feedback stabilizes network dynamics. Together, these results outline how prediction error signaling and top-down feedback pathways shape adaptive sensory processing in biological and artificial systems.

SeminarNeuroscience

AutoMIND: Deep inverse models for revealing neural circuit invariances

Richard Gao
Goethe University
Oct 2, 2025
SeminarNeuroscience

Learning Representations of Complex Meaning in the Human Brain

Leila Wehbe
Associate Professor, Machine Learning Department, Carnegie Mellon University
Feb 24, 2025
SeminarNeuroscienceRecording

On finding what you’re (not) looking for: prospects and challenges for AI-driven discovery

André Curtis Trudel
University of Cincinnati
Oct 10, 2024

Recent high-profile scientific achievements by machine learning (ML) and especially deep learning (DL) systems have reinvigorated interest in ML for automated scientific discovery (eg, Wang et al. 2023). Much of this work is motivated by the thought that DL methods might facilitate the efficient discovery of phenomena, hypotheses, or even models or theories more efficiently than traditional, theory-driven approaches to discovery. This talk considers some of the more specific obstacles to automated, DL-driven discovery in frontier science, focusing on gravitational-wave astrophysics (GWA) as a representative case study. In the first part of the talk, we argue that despite these efforts, prospects for DL-driven discovery in GWA remain uncertain. In the second part, we advocate a shift in focus towards the ways DL can be used to augment or enhance existing discovery methods, and the epistemic virtues and vices associated with these uses. We argue that the primary epistemic virtue of many such uses is to decrease opportunity costs associated with investigating puzzling or anomalous signals, and that the right framework for evaluating these uses comes from philosophical work on pursuitworthiness.

SeminarNeuroscience

Trends in NeuroAI - Brain-like topography in transformers (Topoformer)

Nicholas Blauch
Jun 7, 2024

Dr. Nicholas Blauch will present on his work "Topoformer: Brain-like topographic organization in transformer language models through spatial querying and reweighting". Dr. Blauch is a postdoctoral fellow in the Harvard Vision Lab advised by Talia Konkle and George Alvarez. Paper link: https://openreview.net/pdf?id=3pLMzgoZSA Trends in NeuroAI is a reading group hosted by the MedARC Neuroimaging & AI lab (https://medarc.ai/fmri | https://groups.google.com/g/medarc-fmri).

SeminarNeuroscience

Generative models for video games (rescheduled)

Katja Hoffman
Microsoft Research
May 22, 2024

Developing agents capable of modeling complex environments and human behaviors within them is a key goal of artificial intelligence research. Progress towards this goal has exciting potential for applications in video games, from new tools that empower game developers to realize new creative visions, to enabling new kinds of immersive player experiences. This talk focuses on recent advances of my team at Microsoft Research towards scalable machine learning architectures that effectively capture human gameplay data. In the first part of my talk, I will focus on diffusion models as generative models of human behavior. Previously shown to have impressive image generation capabilities, I present insights that unlock applications to imitation learning for sequential decision making. In the second part of my talk, I discuss a recent project taking ideas from language modeling to build a generative sequence model of an Xbox game.

SeminarNeuroscience

Generative models for video games

Katja Hoffman
Microsoft Research
May 1, 2024

Developing agents capable of modeling complex environments and human behaviors within them is a key goal of artificial intelligence research. Progress towards this goal has exciting potential for applications in video games, from new tools that empower game developers to realize new creative visions, to enabling new kinds of immersive player experiences. This talk focuses on recent advances of my team at Microsoft Research towards scalable machine learning architectures that effectively capture human gameplay data. In the first part of my talk, I will focus on diffusion models as generative models of human behavior. Previously shown to have impressive image generation capabilities, I present insights that unlock applications to imitation learning for sequential decision making. In the second part of my talk, I discuss a recent project taking ideas from language modeling to build a generative sequence model of an Xbox game.

SeminarNeuroscience

Maintaining Plasticity in Neural Networks

Clare Lyle
DeepMind
Mar 13, 2024

Nonstationarity presents a variety of challenges for machine learning systems. One surprising pathology which can arise in nonstationary learning problems is plasticity loss, whereby making progress on new learning objectives becomes more difficult as training progresses. Networks which are unable to adapt in response to changes in their environment experience plateaus or even declines in performance in highly non-stationary domains such as reinforcement learning, where the learner must quickly adapt to new information even after hundreds of millions of optimization steps. The loss of plasticity manifests in a cluster of related empirical phenomena which have been identified by a number of recent works, including the primacy bias, implicit under-parameterization, rank collapse, and capacity loss. While this phenomenon is widely observed, it is still not fully understood. This talk will present exciting recent results which shed light on the mechanisms driving the loss of plasticity in a variety of learning problems and survey methods to maintain network plasticity in non-stationary tasks, with a particular focus on deep reinforcement learning.

SeminarNeuroscience

Trends in NeuroAI - Unified Scalable Neural Decoding (POYO)

Mehdi Azabou
Feb 22, 2024

Lead author Mehdi Azabou will present on his work "POYO-1: A Unified, Scalable Framework for Neural Population Decoding" (https://poyo-brain.github.io/). Mehdi is an ML PhD student at Georgia Tech advised by Dr. Eva Dyer. Paper link: https://arxiv.org/abs/2310.16046 Trends in NeuroAI is a reading group hosted by the MedARC Neuroimaging & AI lab (https://medarc.ai/fmri | https://groups.google.com/g/medarc-fmri).

SeminarNeuroscience

Machine learning for reconstructing, understanding and intervening on neural interactions

Stefano Panzeri
University Medical Center Hamburg-Eppendorf (UKE)
Jan 11, 2024
SeminarNeuroscience

Trends in NeuroAI - Meta's MEG-to-image reconstruction

Reese Kneeland
Jan 5, 2024

Trends in NeuroAI is a reading group hosted by the MedARC Neuroimaging & AI lab (https://medarc.ai/fmri). Title: Brain-optimized inference improves reconstructions of fMRI brain activity Abstract: The release of large datasets and developments in AI have led to dramatic improvements in decoding methods that reconstruct seen images from human brain activity. We evaluate the prospect of further improving recent decoding methods by optimizing for consistency between reconstructions and brain activity during inference. We sample seed reconstructions from a base decoding method, then iteratively refine these reconstructions using a brain-optimized encoding model that maps images to brain activity. At each iteration, we sample a small library of images from an image distribution (a diffusion model) conditioned on a seed reconstruction from the previous iteration. We select those that best approximate the measured brain activity when passed through our encoding model, and use these images for structural guidance during the generation of the small library in the next iteration. We reduce the stochasticity of the image distribution at each iteration, and stop when a criterion on the "width" of the image distribution is met. We show that when this process is applied to recent decoding methods, it outperforms the base decoding method as measured by human raters, a variety of image feature metrics, and alignment to brain activity. These results demonstrate that reconstruction quality can be significantly improved by explicitly aligning decoding distributions to brain activity distributions, even when the seed reconstruction is output from a state-of-the-art decoding algorithm. Interestingly, the rate of refinement varies systematically across visual cortex, with earlier visual areas generally converging more slowly and preferring narrower image distributions, relative to higher-level brain areas. Brain-optimized inference thus offers a succinct and novel method for improving reconstructions and exploring the diversity of representations across visual brain areas. Speaker: Reese Kneeland is a Ph.D. student at the University of Minnesota working in the Naselaris lab. Paper link: https://arxiv.org/abs/2312.07705

SeminarNeuroscience

Trends in NeuroAI - Meta's MEG-to-image reconstruction

Paul Scotti
Dec 7, 2023

Trends in NeuroAI is a reading group hosted by the MedARC Neuroimaging & AI lab (https://medarc.ai/fmri). This will be an informal journal club presentation, we do not have an author of the paper joining us. Title: Brain decoding: toward real-time reconstruction of visual perception Abstract: In the past five years, the use of generative and foundational AI systems has greatly improved the decoding of brain activity. Visual perception, in particular, can now be decoded from functional Magnetic Resonance Imaging (fMRI) with remarkable fidelity. This neuroimaging technique, however, suffers from a limited temporal resolution (≈0.5 Hz) and thus fundamentally constrains its real-time usage. Here, we propose an alternative approach based on magnetoencephalography (MEG), a neuroimaging device capable of measuring brain activity with high temporal resolution (≈5,000 Hz). For this, we develop an MEG decoding model trained with both contrastive and regression objectives and consisting of three modules: i) pretrained embeddings obtained from the image, ii) an MEG module trained end-to-end and iii) a pretrained image generator. Our results are threefold: Firstly, our MEG decoder shows a 7X improvement of image-retrieval over classic linear decoders. Second, late brain responses to images are best decoded with DINOv2, a recent foundational image model. Third, image retrievals and generations both suggest that MEG signals primarily contain high-level visual features, whereas the same approach applied to 7T fMRI also recovers low-level features. Overall, these results provide an important step towards the decoding - in real time - of the visual processes continuously unfolding within the human brain. Speaker: Dr. Paul Scotti (Stability AI, MedARC) Paper link: https://arxiv.org/abs/2310.19812

SeminarNeuroscience

Trends in NeuroAI - SwiFT: Swin 4D fMRI Transformer

Junbeom Kwon
Nov 21, 2023

Trends in NeuroAI is a reading group hosted by the MedARC Neuroimaging & AI lab (https://medarc.ai/fmri). Title: SwiFT: Swin 4D fMRI Transformer Abstract: Modeling spatiotemporal brain dynamics from high-dimensional data, such as functional Magnetic Resonance Imaging (fMRI), is a formidable task in neuroscience. Existing approaches for fMRI analysis utilize hand-crafted features, but the process of feature extraction risks losing essential information in fMRI scans. To address this challenge, we present SwiFT (Swin 4D fMRI Transformer), a Swin Transformer architecture that can learn brain dynamics directly from fMRI volumes in a memory and computation-efficient manner. SwiFT achieves this by implementing a 4D window multi-head self-attention mechanism and absolute positional embeddings. We evaluate SwiFT using multiple large-scale resting-state fMRI datasets, including the Human Connectome Project (HCP), Adolescent Brain Cognitive Development (ABCD), and UK Biobank (UKB) datasets, to predict sex, age, and cognitive intelligence. Our experimental outcomes reveal that SwiFT consistently outperforms recent state-of-the-art models. Furthermore, by leveraging its end-to-end learning capability, we show that contrastive loss-based self-supervised pre-training of SwiFT can enhance performance on downstream tasks. Additionally, we employ an explainable AI method to identify the brain regions associated with sex classification. To our knowledge, SwiFT is the first Swin Transformer architecture to process dimensional spatiotemporal brain functional data in an end-to-end fashion. Our work holds substantial potential in facilitating scalable learning of functional brain imaging in neuroscience research by reducing the hurdles associated with applying Transformer models to high-dimensional fMRI. Speaker: Junbeom Kwon is a research associate working in Prof. Jiook Cha’s lab at Seoul National University. Paper link: https://arxiv.org/abs/2307.05916

SeminarNeuroscience

BrainLM Journal Club

Connor Lane
Sep 29, 2023

Connor Lane will lead a journal club on the recent BrainLM preprint, a foundation model for fMRI trained using self-supervised masked autoencoder training. Preprint: https://www.biorxiv.org/content/10.1101/2023.09.12.557460v1 Tweeprint: https://twitter.com/david_van_dijk/status/1702336882301112631?t=Q2-U92-BpJUBh9C35iUbUA&s=19

SeminarNeuroscience

Algonauts 2023 winning paper journal club (fMRI encoding models)

Huzheng Yang, Paul Scotti
Aug 18, 2023

Algonauts 2023 was a challenge to create the best model that predicts fMRI brain activity given a seen image. Huze team dominated the competition and released a preprint detailing their process. This journal club meeting will involve open discussion of the paper with Q/A with Huze. Paper: https://arxiv.org/pdf/2308.01175.pdf Related paper also from Huze that we can discuss: https://arxiv.org/pdf/2307.14021.pdf

SeminarNeuroscience

1.8 billion regressions to predict fMRI (journal club)

Mihir Tripathy
Jul 28, 2023

Public journal club where this week Mihir will present on the 1.8 billion regressions paper (https://www.biorxiv.org/content/10.1101/2022.03.28.485868v2), where the authors use hundreds of pretrained model embeddings to best predict fMRI activity.

SeminarNeuroscienceRecording

In search of the unknown: Artificial intelligence and foraging

Nathan Wispinski & Paulo Bruno Serafim
University of Alberta & Gran Sasso Science Institute
Jul 11, 2023
SeminarNeuroscience

Decoding mental conflict between reward and curiosity in decision-making

Naoki Honda
Hiroshima University
Jul 11, 2023

Humans and animals are not always rational. They not only rationally exploit rewards but also explore an environment owing to their curiosity. However, the mechanism of such curiosity-driven irrational behavior is largely unknown. Here, we developed a decision-making model for a two-choice task based on the free energy principle, which is a theory integrating recognition and action selection. The model describes irrational behaviors depending on the curiosity level. We also proposed a machine learning method to decode temporal curiosity from behavioral data. By applying it to rat behavioral data, we found that the rat had negative curiosity, reflecting conservative selection sticking to more certain options and that the level of curiosity was upregulated by the expected future information obtained from an uncertain environment. Our decoding approach can be a fundamental tool for identifying the neural basis for reward–curiosity conflicts. Furthermore, it could be effective in diagnosing mental disorders.

SeminarNeuroscience

Relations and Predictions in Brains and Machines

Kim Stachenfeld
Deepmind
Apr 7, 2023

Humans and animals learn and plan with flexibility and efficiency well beyond that of modern Machine Learning methods. This is hypothesized to owe in part to the ability of animals to build structured representations of their environments, and modulate these representations to rapidly adapt to new settings. In the first part of this talk, I will discuss theoretical work describing how learned representations in hippocampus enable rapid adaptation to new goals by learning predictive representations, while entorhinal cortex compresses these predictive representations with spectral methods that support smooth generalization among related states. I will also cover recent work extending this account, in which we show how the predictive model can be adapted to the probabilistic setting to describe a broader array of generalization results in humans and animals, and how entorhinal representations can be modulated to support sample generation optimized for different behavioral states. In the second part of the talk, I will overview some of the ways in which we have combined many of the same mathematical concepts with state-of-the-art deep learning methods to improve efficiency and performance in machine learning applications like physical simulation, relational reasoning, and design.

SeminarNeuroscience

Bridging machine learning and mechanistic modelling

Jakob Macke
University of Tubingen
Mar 15, 2023
SeminarNeuroscienceRecording

Understanding Machine Learning via Exactly Solvable Statistical Physics Models

Lenka Zdeborová
EPFL
Feb 8, 2023

The affinity between statistical physics and machine learning has a long history. I will describe the main lines of this long-lasting friendship in the context of current theoretical challenges and open questions about deep learning. Theoretical physics often proceeds in terms of solvable synthetic models, I will describe the related line of work on solvable models of simple feed-forward neural networks. I will highlight a path forward to capture the subtle interplay between the structure of the data, the architecture of the network, and the optimization algorithms commonly used for learning.

SeminarNeuroscience

Maths, AI and Neuroscience Meeting Stockholm

Roshan Cools, Alain Destexhe, Upi Bhalla, Vijay Balasubramnian, Dinos Meletis, Richard Naud
Dec 15, 2022

To understand brain function and develop artificial general intelligence it has become abundantly clear that there should be a close interaction among Neuroscience, machine learning and mathematics. There is a general hope that understanding the brain function will provide us with more powerful machine learning algorithms. On the other hand advances in machine learning are now providing the much needed tools to not only analyse brain activity data but also to design better experiments to expose brain function. Both neuroscience and machine learning explicitly or implicitly deal with high dimensional data and systems. Mathematics can provide powerful new tools to understand and quantify the dynamics of biological and artificial systems as they generate behavior that may be perceived as intelligent.

SeminarNeuroscience

Experimental Neuroscience Bootcamp

Adam Kampff
Voight Kampff, London, UK
Dec 5, 2022

This course provides a fundamental foundation in the modern techniques of experimental neuroscience. It introduces the essentials of sensors, motor control, microcontrollers, programming, data analysis, and machine learning by guiding students through the “hands on” construction of an increasingly capable robot. In parallel, related concepts in neuroscience are introduced as nature’s solution to the challenges students encounter while designing and building their own intelligent system.

SeminarNeuroscienceRecording

On the link between conscious function and general intelligence in humans and machines

Arthur Juliani
Microsoft Research
Nov 18, 2022

In popular media, there is often a connection drawn between the advent of awareness in artificial agents and those same agents simultaneously achieving human or superhuman level intelligence. In this talk, I will examine the validity and potential application of this seemingly intuitive link between consciousness and intelligence. I will do so by examining the cognitive abilities associated with three contemporary theories of conscious function: Global Workspace Theory (GWT), Information Generation Theory (IGT), and Attention Schema Theory (AST), and demonstrating that all three theories specifically relate conscious function to some aspect of domain-general intelligence in humans. With this insight, we will turn to the field of Artificial Intelligence (AI) and find that, while still far from demonstrating general intelligence, many state-of-the-art deep learning methods have begun to incorporate key aspects of each of the three functional theories. Given this apparent trend, I will use the motivating example of mental time travel in humans to propose ways in which insights from each of the three theories may be combined into a unified model. I believe that doing so can enable the development of artificial agents which are not only more generally intelligent but are also consistent with multiple current theories of conscious function.

SeminarNeuroscienceRecording

From Machine Learning to Autonomous Intelligence

Yann Le Cun
Meta-FAIR & Meta AI
Oct 19, 2022

How could machines learn as efficiently as humans and animals? How could machines learn to reason and plan? How could machines learn representations of percepts and action plans at multiple levels of abstraction, enabling them to reason, predict, and plan at multiple time horizons? I will propose a possible path towards autonomous intelligent agents, based on a new modular cognitive architecture and a somewhat new self supervised training paradigm. The centerpiece of the proposed architecture is a configurable predictive world model that allows the agent to plan. Behavior and learning are driven by a set of differentiable intrinsic cost functions. The world model uses a new type of energy-based model architecture called H-JEPA (Hierarchical Joint Embedding Predictive Architecture). H-JEPA learns hierarchical abstract representations of the world that are simultaneously maximally informative and maximally predictable.

SeminarNeuroscienceRecording

Learning Relational Rules from Rewards

Guillermo Puebla
University of Bristol
Oct 13, 2022

Humans perceive the world in terms of objects and relations between them. In fact, for any given pair of objects, there is a myriad of relations that apply to them. How does the cognitive system learn which relations are useful to characterize the task at hand? And how can it use these representations to build a relational policy to interact effectively with the environment? In this paper we propose that this problem can be understood through the lens of a sub-field of symbolic machine learning called relational reinforcement learning (RRL). To demonstrate the potential of our approach, we build a simple model of relational policy learning based on a function approximator developed in RRL. We trained and tested our model in three Atari games that required to consider an increasingly number of potential relations: Breakout, Pong and Demon Attack. In each game, our model was able to select adequate relational representations and build a relational policy incrementally. We discuss the relationship between our model with models of relational and analogical reasoning, as well as its limitations and future directions of research.

SeminarNeuroscience

From Machine Learning to Autonomous Intelligence

Yann LeCun
Meta Fair
Oct 10, 2022

How could machines learn as efficiently as humans and animals? How could machines learn to reason and plan? How could machines learn representations of percepts and action plans at multiple levels of abstraction, enabling them to reason, predict, and plan at multiple time horizons? I will propose a possible path towards autonomous intelligent agents, based on a new modular cognitive architecture and a somewhat new self-supervised training paradigm. The centerpiece of the proposed architecture is a configurable predictive world model that allows the agent to plan. Behavior and learning are driven by a set of differentiable intrinsic cost functions. The world model uses a new type of energy-based model architecture called H-JEPA (Hierarchical Joint Embedding Predictive Architecture). H-JEPA learns hierarchical abstract representations of the world that are simultaneously maximally informative and maximally predictable. The corresponding working paper is available here:https://openreview.net/forum?id=BZ5a1r-kVsf

SeminarNeuroscienceRecording

Spontaneous Emergence of Computation in Network Cascades

Galen Wilkerson
Imperial College London
Aug 6, 2022

Neuronal network computation and computation by avalanche supporting networks are of interest to the fields of physics, computer science (computation theory as well as statistical or machine learning) and neuroscience. Here we show that computation of complex Boolean functions arises spontaneously in threshold networks as a function of connectivity and antagonism (inhibition), computed by logic automata (motifs) in the form of computational cascades. We explain the emergent inverse relationship between the computational complexity of the motifs and their rank-ordering by function probabilities due to motifs, and its relationship to symmetry in function space. We also show that the optimal fraction of inhibition observed here supports results in computational neuroscience, relating to optimal information processing.

SeminarNeuroscience

Attention in Psychology, Neuroscience, and Machine Learning

Grace Lindsay
NYU
Jun 15, 2022
SeminarNeuroscienceRecording

Canonical neural networks perform active inference

Takuya Isomura
RIKEN CBS
Jun 10, 2022

The free-energy principle and active inference have received a significant attention in the fields of neuroscience and machine learning. However, it remains to be established whether active inference is an apt explanation for any given neural network that actively exchanges with its environment. To address this issue, we show that a class of canonical neural networks of rate coding models implicitly performs variational Bayesian inference under a well-known form of partially observed Markov decision process model (Isomura, Shimazaki, Friston, Commun Biol, 2022). Based on the proposed theory, we demonstrate that canonical neural networks—featuring delayed modulation of Hebbian plasticity—can perform planning and adaptive behavioural control in the Bayes optimal manner, through postdiction of their previous decisions. This scheme enables us to estimate implicit priors under which the agent’s neural network operates and identify a specific form of the generative model. The proposed equivalence is crucial for rendering brain activity explainable to better understand basic neuropsychology and psychiatric disorders. Moreover, this notion can dramatically reduce the complexity of designing self-learning neuromorphic hardware to perform various types of tasks.

SeminarNeuroscienceRecording

Hebbian Plasticity Supports Predictive Self-Supervised Learning of Disentangled Representations​

Manu Halvagal​
Friedrich Miescher Institute for Biomedical Research
May 4, 2022

Discriminating distinct objects and concepts from sensory stimuli is essential for survival. Our brains accomplish this feat by forming meaningful internal representations in deep sensory networks with plastic synaptic connections. Experience-dependent plasticity presumably exploits temporal contingencies between sensory inputs to build these internal representations. However, the precise mechanisms underlying plasticity remain elusive. We derive a local synaptic plasticity model inspired by self-supervised machine learning techniques that shares a deep conceptual connection to Bienenstock-Cooper-Munro (BCM) theory and is consistent with experimentally observed plasticity rules. We show that our plasticity model yields disentangled object representations in deep neural networks without the need for supervision and implausible negative examples. In response to altered visual experience, our model qualitatively captures neuronal selectivity changes observed in the monkey inferotemporal cortex in-vivo. Our work suggests a plausible learning rule to drive learning in sensory networks while making concrete testable predictions.

SeminarNeuroscienceRecording

Population coding in the cerebellum: a machine learning perspective

Reza Shadmehr
Johns Hopkins School of Medicine
Apr 6, 2022

The cerebellum resembles a feedforward, three-layer network of neurons in which the “hidden layer” consists of Purkinje cells (P-cells) and the output layer consists of deep cerebellar nucleus (DCN) neurons. In this analogy, the output of each DCN neuron is a prediction that is compared with the actual observation, resulting in an error signal that originates in the inferior olive. Efficient learning requires that the error signal reach the DCN neurons, as well as the P-cells that project onto them. However, this basic rule of learning is violated in the cerebellum: the olivary projections to the DCN are weak, particularly in adulthood. Instead, an extraordinarily strong signal is sent from the olive to the P-cells, producing complex spikes. Curiously, P-cells are grouped into small populations that converge onto single DCN neurons. Why are the P-cells organized in this way, and what is the membership criterion of each population? Here, I apply elementary mathematics from machine learning and consider the fact that P-cells that form a population exhibit a special property: they can synchronize their complex spikes, which in turn suppress activity of DCN neuron they project to. Thus complex spikes cannot only act as a teaching signal for a P-cell, but through complex spike synchrony, a P-cell population may act as a surrogate teacher for the DCN neuron that produced the erroneous output. It appears that grouping of P-cells into small populations that share a preference for error satisfies a critical requirement of efficient learning: providing error information to the output layer neuron (DCN) that was responsible for the error, as well as the hidden layer neurons (P-cells) that contributed to it. This population coding may account for several remarkable features of behavior during learning, including multiple timescales, protection from erasure, and spontaneous recovery of memory.

SeminarNeuroscienceRecording

CNStalk: Using machine learning to predict mental health on the basis of brain, behaviour and environment

Andre Marquand
Donders Institute
Mar 31, 2022
SeminarNeuroscience

Interdisciplinary College

Tarek Besold, Suzanne Dikker, Astrid Prinz, Fynn-Mathis Trautwein, Niklas Keller, Ida Momennejad, Georg von Wichert
Mar 7, 2022

The Interdisciplinary College is an annual spring school which offers a dense state-of-the-art course program in neurobiology, neural computation, cognitive science/psychology, artificial intelligence, machine learning, robotics and philosophy. It is aimed at students, postgraduates and researchers from academia and industry. This year's focus theme "Flexibility" covers (but not be limited to) the nervous system, the mind, communication, and AI & robotics. All this will be packed into a rich, interdisciplinary program of single- and multi-lecture courses, and less traditional formats.

SeminarNeuroscienceRecording

Implementing structure mapping as a prior in deep learning models for abstract reasoning

Shashank Shekhar
University of Guelph
Mar 3, 2022

Building conceptual abstractions from sensory information and then reasoning about them is central to human intelligence. Abstract reasoning both relies on, and is facilitated by, our ability to make analogies about concepts from known domains to novel domains. Structure Mapping Theory of human analogical reasoning posits that analogical mappings rely on (higher-order) relations and not on the sensory content of the domain. This enables humans to reason systematically about novel domains, a problem with which machine learning (ML) models tend to struggle. We introduce a two-stage neural net framework, which we label Neural Structure Mapping (NSM), to learn visual analogies from Raven's Progressive Matrices, an abstract visual reasoning test of fluid intelligence. Our framework uses (1) a multi-task visual relationship encoder to extract constituent concepts from raw visual input in the source domain, and (2) a neural module net analogy inference engine to reason compositionally about the inferred relation in the target domain. Our NSM approach (a) isolates the relational structure from the source domain with high accuracy, and (b) successfully utilizes this structure for analogical reasoning in the target domain.

SeminarNeuroscienceRecording

Taming chaos in neural circuits

Rainer Engelken
Columbia University
Feb 23, 2022

Neural circuits exhibit complex activity patterns, both spontaneously and in response to external stimuli. Information encoding and learning in neural circuits depend on the ability of time-varying stimuli to control spontaneous network activity. In particular, variability arising from the sensitivity to initial conditions of recurrent cortical circuits can limit the information conveyed about the sensory input. Spiking and firing rate network models can exhibit such sensitivity to initial conditions that are reflected in their dynamic entropy rate and attractor dimensionality computed from their full Lyapunov spectrum. I will show how chaos in both spiking and rate networks depends on biophysical properties of neurons and the statistics of time-varying stimuli. In spiking networks, increasing the input rate or coupling strength aids in controlling the driven target circuit, which is reflected in both a reduced trial-to-trial variability and a decreased dynamic entropy rate. With sufficiently strong input, a transition towards complete network state control occurs. Surprisingly, this transition does not coincide with the transition from chaos to stability but occurs at even larger values of external input strength. Controllability of spiking activity is facilitated when neurons in the target circuit have a sharp spike onset, thus a high speed by which neurons launch into the action potential. I will also discuss chaos and controllability in firing-rate networks in the balanced state. For these, external control of recurrent dynamics strongly depends on correlations in the input. This phenomenon was studied with a non-stationary dynamic mean-field theory that determines how the activity statistics and the largest Lyapunov exponent depend on frequency and amplitude of the input, recurrent coupling strength, and network size. This shows that uncorrelated inputs facilitate learning in balanced networks. The results highlight the potential of Lyapunov spectrum analysis as a diagnostic for machine learning applications of recurrent networks. They are also relevant in light of recent advances in optogenetics that allow for time-dependent stimulation of a select population of neurons.

SeminarNeuroscience

Machine learning for measuring and modeling the motor system

Mackenzie Mathis
EPFL
Feb 16, 2022

machine learning coverage

60 items

Seminar50
Position10
Domain spotlight

Explore how machine learning research is advancing inside Neuro.

Visit domain