Scalable Reinforcement Learning Policies for Multi-Agent Control

This paper develops a stochastic Multi-Agent Reinforcement Learning (MARL) method to learn control policies that can handle an arbitrary number of external agents; our policies can be executed for tasks consisting of 1000 pursuers and 1000 evaders. We model pursuers as agents with limited on-board sensing and formulate the problem as a decentralized, partially-observable Markov Decision Process. An attention mechanism is used to build a permutation and input-size invariant embedding of the observations for learning a stochastic policy and value function using techniques in entropy-regularized off-policy methods. Simulation experiments on a large number of problems show that our control policies are dramatically scalable and display cooperative behavior in spite of being executed in a decentralized fashion; our methods offer a simple solution to classical multi-agent problems using techniques in reinforcement learning.

An Information-Geometric Distance on the Space of Tasks

This paper computes a distance between tasks modeled as joint distributions on data and labels. We develop a stochastic process that transports the marginal on the data of the source task to that of the target task, and simultaneously updates the weights of a classifier initialized on the source task to track this evolving data distribution. The distance between two tasks is defined to be the shortest path on the Riemannian manifold of the conditional distribution of labels given data as the weights evolve. We derive connections of this distance with Rademacher complexity-based generalization bounds; distance between tasks computed using our method can be interpreted as the trajectory in weight space that keeps the generalization gap constant as the task distribution changes from the source to the target. Experiments on image classification datasets show that this task distance helps predict the performance of transfer learning: fine-tuning techniques have an easier time transferring to tasks that are close to each other under our distance.

MIDAS: Multi-agent Interaction-aware Decision-making with Adaptive Strategies for Urban Autonomous Navigation

Autonomous navigation in crowded, complex urban environments requires interacting with other agents on the road. A common solution to this problem is to use a prediction model to guess the likely future actions of other agents. While this is reasonable, it leads to overly conservative plans because it does not explicitly model the mutual influence of the actions of interacting agents. This paper builds a reinforcement learning-based method named MIDAS where an ego-agent learns to affect the control actions of other cars in urban driving scenarios. MIDAS uses an attention-mechanism to handle an arbitrary number of other agents and includes a ''driver-type'' parameter to learn a single policy that works across different planning objectives. We build a simulation environment that enables diverse interaction experiments with a large number of agents and methods for quantitatively studying the safety, efficiency, and interaction among vehicles. MIDAS is validated using extensive experiments and we show that it (i) can work across different road geometries, (ii) results in an adaptive ego policy that can be tuned easily to satisfy performance criteria such as aggressive or cautious driving, (iii) is robust to changes in the driving policies of external agents, and (iv) is more efficient and safer than existing approaches to interaction-aware decision-making.

Proximal Deterministic Policy Gradient

This paper introduces two simple techniques to improve off-policy Reinforcement Learning (RL) algorithms. First, we formulate off-policy RL as a stochastic proximal point iteration. The target network plays the role of the variable of optimization and the value network computes the proximal operator. Second, we exploits the two value functions commonly employed in state-of-the-art off-policy algorithms to provide an improved action value estimate through bootstrapping with limited increase of computational resources. Further, we demonstrate significant performance improvement over state-of-the-art algorithms on standard continuous-control RL benchmarks.

DDPG++: Striving for Simplicity in Continuous-control Off-Policy Reinforcement Learning

This paper prescribes a suite of techniques for off-policy Reinforcement Learning (RL) that simplify the training process and reduce the sample complexity. First, we show that simple Deterministic Policy Gradient works remarkably well as long as the overestimation bias is controlled. This is contrast to existing literature which creates sophisticated off-policy techniques. Second, we pinpoint training instabilities, typical of off-policy algorithms, to the greedy policy update step; existing solutions such as delayed policy updates do not mitigate this issue. Third, we show that ideas in the propensity estimation literature can be used to importance-sample transitions from the replay buffer and selectively update the policy to prevent deterioration of performance. We make these claims using extensive experimentation on a set of challenging MuJoCo tasks. A short video of our results can be seen at https://tinyurl.com/scs6p5m .

Fast, Accurate, and Simple Models for Tabular Data via Augmented Distillation

Automated machine learning (AutoML) can produce complex model ensembles by stacking, bagging, and boosting many individual models like trees, deep networks, and nearest neighbor estimators. While highly accurate, the resulting predictors are large, slow, and opaque as compared to their constituents. To improve the deployment of AutoML on tabular data, we propose FAST-DAD to distill arbitrarily complex ensemble predictors into individual models like boosted trees, random forests, and deep networks. At the heart of our approach is a data augmentation strategy based on Gibbs sampling from a self-attention pseudolikelihood estimator. Across 30 datasets spanning regression and binary/multiclass classification tasks, FAST-DAD distillation produces significantly better individual models than one obtains through standard training on the original data. Our individual distilled models are over 10x faster and more accurate than ensemble predictors produced by AutoML tools like H2O/AutoSklearn.

BayesRace: Learning to race autonomously using prior experience

Learning to race autonomously is a challenging problem. It requires perception, estimation, planning, and control to work together in synchronization while driving at the limit of a vehicle's handling capability. Among others, one of the fundamental challenges lies in predicting the vehicle's future states like position, orientation, and speed with high accuracy because it is inevitably hard to identify vehicle model parameters that capture its real nonlinear dynamics in the presence of lateral tire slip. We present a model-based planning and control framework for autonomous racing that significantly reduces the effort required in system identification. Our approach bridges the gap between the design in a simulation and the real world by learning from on-board sensor measurements. Thus, the teams participating in autonomous racing competitions can start racing on new tracks without having to worry about tuning the vehicle model.

TraDE: Transformers for Density Estimation

We present TraDE, an attention-based architecture for auto-regressive density estimation. In addition to a Maximum Likelihood loss we employ a Maximum Mean Discrepancy (MMD) two-sample loss to ensure that samples from the estimate resemble the training data. The use of attention means that the model need not retain conditional sufficient statistics during the process beyond what is needed for each covariate. TraDE performs significantly better than existing approaches such differentiable flow based estimators on standard tabular and image-based benchmarks in terms of the log-likelihood on held out data. TraDE works well wide range of tasks that includes classification methods to ascertain the quality of generated samples, out of distribution sample detection, and handling outliers in the training data.

A Free-Energy Principle for Representation Learning

This paper employs a formal connection of machine learning with thermodynamics to characterize the quality of learnt representations for transfer learning. We discuss how information-theoretic functional such as rate, distortion and classification loss of a model lie on a convex, so-called equilibrium surface.We prescribe dynamical processes to traverse this surface under constraints, e.g., an iso-classification process that trades off rate and distortion to keep the classification loss unchanged. We demonstrate how this process can be used for transferring representations from a source dataset to a target dataset while keeping the classification loss constant. Experimental validation of the theoretical results is provided on standard image-classification datasets.

Rethinking the Hyperparameters for Fine-tuning

Fine-tuning from pre-trained ImageNet models has become the de-facto standard for various computer vision tasks. Current practices for fine-tuning typically involve selecting an ad-hoc choice of hyperparameters and keeping them fixed to values normally used for training from scratch. This paper re-examines several common practices of setting hyperparameters for fine-tuning. Our findings are based on extensive empirical evaluation for fine-tuning on various transfer learning benchmarks. (1) While prior works have thoroughly investigated learning rate and batch size, momentum for fine-tuning is a relatively unexplored parameter. We find that the value of momentum also affects fine-tuning performance and connect it with previous theoretical findings. (2) Optimal hyperparameters for fine-tuning, in particular, the effective learning rate, are not only dataset dependent but also sensitive to the similarity between the source domain and target domain. This is in contrast to hyperparameters for training from scratch. (3) Reference-based regularization that keeps models close to the initial model does not necessarily apply for "dissimilar" datasets. Our findings challenge common practices of fine-tuning and encourages deep learning practitioners to rethink the hyperparameters for fine-tuning.