ترغب بنشر مسار تعليمي؟ اضغط هنا

Neural Replicator Dynamics

138   0   0.0 ( 0 )
 نشر من قبل Shayegan Omidshafiei
 تاريخ النشر 2019
  مجال البحث الهندسة المعلوماتية
والبحث باللغة English




اسأل ChatGPT حول البحث

Policy gradient and actor-critic algorithms form the basis of many commonly used training techniques in deep reinforcement learning. Using these algorithms in multiagent environments poses problems such as nonstationarity and instability. In this paper, we first demonstrate that standard softmax-based policy gradient can be prone to poor performance in the presence of even the most benign nonstationarity. By contrast, it is known that the replicator dynamics, a well-studied model from evolutionary game theory, eliminates dominated strategies and exhibits convergence of the time-averaged trajectories to interior Nash equilibria in zero-sum games. Thus, using the replicator dynamics as a foundation, we derive an elegant one-line change to policy gradient methods that simply bypasses the gradient step through the softmax, yielding a new algorithm titled Neural Replicator Dynamics (NeuRD). NeuRD reduces to the exponential weights/Hedge algorithm in the single-state all-actions case. Additionally, NeuRD has formal equivalence to softmax counterfactual regret minimization, which guarantees convergence in the sequential tabular case. Importantly, our algorithm provides a straightforward way of extending the replicator dynamics to the function approximation setting. Empirical results show that NeuRD quickly adapts to nonstationarities, outperforming policy gradient significantly in both tabular and function approximation settings, when evaluated on the standard imperfect information benchmarks of Kuhn Poker, Leduc Poker, and Goofspiel.



قيم البحث

اقرأ أيضاً

We describe TF-Replicator, a framework for distributed machine learning designed for DeepMind researchers and implemented as an abstraction over TensorFlow. TF-Replicator simplifies writing data-parallel and model-parallel research code. The same mod els can be effortlessly deployed to different cluster architectures (i.e. one or many machines containing CPUs, GPUs or TPU accelerators) using synchronous or asynchronous training regimes. To demonstrate the generality and scalability of TF-Replicator, we implement and benchmark three very different models: (1) A ResNet-50 for ImageNet classification, (2) a SN-GAN for class-conditional ImageNet image generation, and (3) a D4PG reinforcement learning agent for continuous control. Our results show strong scalability performance without demanding any distributed systems expertise of the user. The TF-Replicator programming model will be open-sourced as part of TensorFlow 2.0 (see https://github.com/tensorflow/community/pull/25).
415 - Qi She , Anqi Wu 2019
Latent dynamics discovery is challenging in extracting complex dynamics from high-dimensional noisy neural data. Many dimensionality reduction methods have been widely adopted to extract low-dimensional, smooth and time-evolving latent trajectories. However, simple state transition structures, linear embedding assumptions, or inflexible inference networks impede the accurate recovery of dynamic portraits. In this paper, we propose a novel latent dynamic model that is capable of capturing nonlinear, non-Markovian, long short-term time-dependent dynamics via recurrent neural networks and tackling complex nonlinear embedding via non-parametric Gaussian process. Due to the complexity and intractability of the model and its inference, we also provide a powerful inference network with bi-directional long short-term memory networks that encode both past and future information into posterior distributions. In the experiment, we show that our model outperforms other state-of-the-art methods in reconstructing insightful latent dynamics from both simulated and experimental neural datasets with either Gaussian or Poisson observations, especially in the low-sample scenario. Our codes and additional materials are available at https://github.com/sheqi/GP-RNN_UAI2019.
Deep neural networks (DNNs) are powerful black-box predictors that have achieved impressive performance on a wide variety of tasks. However, their accuracy comes at the cost of intelligibility: it is usually unclear how they make their decisions. Thi s hinders their applicability to high stakes decision-making domains such as healthcare. We propose Neural Additive Models (NAMs) which combine some of the expressivity of DNNs with the inherent intelligibility of generalized additive models. NAMs learn a linear combination of neural networks that each attend to a single input feature. These networks are trained jointly and can learn arbitrarily complex relationships between their input feature and the output. Our experiments on regression and classification datasets show that NAMs are more accurate than widely used intelligible models such as logistic regression and shallow decision trees. They perform similarly to existing state-of-the-art generalized additive models in accuracy, but can be more easily applied to real-world problems.
We introduce the textit{epistemic neural network} (ENN) as an interface for uncertainty modeling in deep learning. All existing approaches to uncertainty modeling can be expressed as ENNs, and any ENN can be identified with a Bayesian neural network. However, this new perspective provides several promising directions for future research. Where prior work has developed probabilistic inference tools for neural networks; we ask instead, `which neural networks are suitable as tools for probabilistic inference?. We propose a clear and simple metric for progress in ENNs: the KL-divergence with respect to a target distribution. We develop a computational testbed based on inference in a neural network Gaussian process and release our code as a benchmark at url{https://github.com/deepmind/enn}. We evaluate several canonical approaches to uncertainty modeling in deep learning, and find they vary greatly in their performance. We provide insight to the sensitivity of these results and show that our metric is highly correlated with performance in sequential decision problems. Finally, we provide indications that new ENN architectures can improve performance in both the statistical quality and computational cost.
We present a new non-Archimedean model of evolutionary dynamics, in which the genomes are represented by p-adic numbers. In this model the genomes have a variable length, not necessarily bounded, in contrast with the classical models where the length is fixed. The time evolution of the concentration of a given genome is controlled by a p-adic evolution equation. This equation depends on a fitness function f and on mutation measure Q. By choosing a mutation measure of Gibbs type, and by using a p-adic version of the Maynard Smith Ansatz, we show the existence of threshold function M_{c}(f,Q), such that the long term survival of a genome requires that its length grows faster than M_{c}(f,Q). This implies that Eigens paradox does not occur if the complexity of genomes grows at the right pace. About twenty years ago, Scheuring and Poole, Jeffares, Penny proposed a hypothesis to explain Eigens paradox. Our mathematical model shows that this biological hypothesis is feasible, but it requires p-adic analysis instead of real analysis. More exactly, the Darwin-Eigen cycle proposed by Poole et al. takes place if the length of the genomes exceeds M_{c}(f,Q).

الأسئلة المقترحة

التعليقات
جاري جلب التعليقات جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
mircosoft-partner

هل ترغب بارسال اشعارات عن اخر التحديثات في شمرا-اكاديميا