Multi-Agent Reinforcement Learning (MARL) provides a powerful framework for learning coordination in multi-agent systems. However, applying MARL to robotics still remains challenging due to high-dimensional continuous joint action spaces, complex reward design, and non-stationary transitions inherent to decentralized settings. On the other hand, humans learn complex coordination through staged curricula, where long-horizon behaviors are progressively built upon simpler skills. Motivated by this, we propose CRAFT: Coaching Reinforcement learning Autonomously using Foundation models for multi-robot coordination Tasks, a framework that leverages the reasoning capabilities of foundation models to act as a "coach" for multi-robot coordination. CRAFT automatically decomposes long-horizon coordination tasks into sequences of subtasks using the planning capability of Large Language Models (LLMs). In what follows, CRAFT trains each subtask using reward functions generated by LLM, and refines them through a Vision Language Model (VLM)-guided reward-refinement loop. We evaluate CRAFT on multi-quadruped navigation and bimanual manipulation tasks, demonstrating its capability to learn complex coordination behaviors. In addition, we validate the multi-quadruped navigation policy in real hardware experiments.

Animals flexibly recombine a finite set of core motor motifs to meet diverse task demands, but existing behavior segmentation methods oversimplify this process by imposing discrete syllables under restrictive generative assumptions. To better capture the continuous structure of behavior generation, we introduce motif-based continuous dynamics (MCD) discovery, a framework that (1) uncovers interpretable motif sets as latent basis functions of behavior by leveraging representations of behavioral transition structure, and (2) models behavioral dynamics as continuously evolving mixtures of these motifs. We validate MCD on a multi-task gridworld, a labyrinth navigation task, and freely moving animal behavior. Across settings, it identifies reusable motif components, captures continuous compositional dynamics, and generates realistic trajectories beyond the capabilities of traditional discrete segmentation models. By providing a generative account of how complex animal behaviors emerge from dynamic combinations of fundamental motor motifs, our approach advances the quantitative study of natural behavior.

For the analysis of the Track-and-Stop strategy we derive a novel concentration inequality for Markov chains that may be of interest in its own right.

In this work, we propose a novel type of normalizing flow driven by a differential deformation of the Wiener process.

Neural Information Processing Systems http://nips.cc/

"NIPS Neural Information Processing Systems 8-11th December 2014, Montreal, Canada",,, "Paper ID:","1694" "Title:","An Integer Polynomial Programming Based Framework for Lifted MAP Inference" Current Reviews First provide a summary of the paper, and then address the following criteria: Quality, clarity, originality and significance. The authors propose a new method of performing lifted inference on Markov Logic Networks. The essence of the idea is to encode the MLN as an integer polynomial program, which is then transformed into an integer linear program (which could be solved with a conventional solver such as Gurobi or CPlex). The lifting as preprocessing idea due to Sarkhel et al. is appealing, and this paper extends it using ideas from probabilistic theorem proving. The idea is to extend the list of symmetries recognized by this lifted inference approach (Algorithm 1).

Your skepticism of growBuf may stem from a shortcoming in our presentation of the method. For most real-world applications, the memory in the chain tends to be fairly local (non-unitary second eigenvalue of the transition matrix), especially in the context of our long chains of interest.

V ariational inference algorithms have proven successful for Bayesian analysis in large data settings, with recent advances using stochastic variational inference (SVI). However, such methods have largely been studied in independent or exchangeable data settings. We develop an SVI algorithm to learn the parameters of hidden Markov models (HMMs) in a time-dependent data setting. The challenge in applying stochastic optimization in this setting arises from dependencies in the chain, which must be broken to consider minibatches of observations. We propose an algorithm that harnesses the memory decay of the chain to adaptively bound errors arising from edge effects. We demonstrate the effectiveness of our algorithm on synthetic experiments and a large genomics dataset where a batch algorithm is computationally infeasible.

Neural population activity in cortical circuits is not solely driven by external inputs, but is also modulated by endogenous states which vary on multiple time-scales. To understand information processing in cortical circuits, we need to understand the statistical structure of internal states and their interaction with sensory inputs. Here, we present a statistical model for extracting hierarchically organised neural population states from multi-channel recordings of neural spiking activity. Population states are modelled using a hidden Markov decision tree with state-dependent tuning parameters and a generalised linear observation model. We present a varia-tional Bayesian inference algorithm for estimating the posterior distribution over parameters from neural population recordings. On simulated data, we show that we can identify the underlying sequence of population states and reconstruct the ground truth parameters. Using population recordings from visual cortex, we find that a model with two levels of population states outperforms both a one-state and a two-state generalised linear model. Finally, we find that modelling of state-dependence also improves the accuracy with which sensory stimuli can be decoded from the population response.