Shaun Patel has such a tranquil voice that it's easy to see how he convinces patients to let him experiment in the depth of their brains. On the phone, in his office at Massachusetts General Hospital (he is also on faculty at Harvard Medical School), the neuroscientist spoke about gray matter almost as if he were guiding me in meditation. Or perhaps that was just the heady effect of him detailing a paper he had just published in Brain, showing how, using implants on his patients, he could enhance learning by stimulating the caudate nucleus, which lies near the center of the brain.1 You have to time the electric pulse just right, he told me, based on the activity of certain neurons firing during an active learning phase of a game. A perfectly timed pulse could speed up how quickly his patients made the right associations. Using similar methods, he said he has induced people to make more financially conservative bets.

Meta-Reinforcement learning approaches aim to develop learning procedures that can adapt quickly to a distribution of tasks with the help of a few examples. Developing efficient exploration strategies capable of finding the most useful samples becomes critical in such settings. Existing approaches towards finding efficient exploration strategies add auxiliary objectives to promote exploration by the pre-update policy, however, this makes the adaptation using a few gradient steps difficult as the pre-update (exploration) and post-update (exploitation) policies are often quite different. Instead, we propose to explicitly model a separate exploration policy for the task distribution. Having two different policies gives more flexibility in training the exploration policy and also makes adaptation to any specific task easier. We show that using self-supervised or supervised learning objectives for adaptation allows for more efficient inner-loop updates and also demonstrate the superior performance of our model compared to prior works in this domain.

Detecting earthquake events from seismic time series has proved itself a challenging task. Manual detection can be expensive and tedious due to the intensive labor and large scale data set. In recent years, automatic detection methods based on machine learning have been developed to improve accuracy and efficiency. However, the accuracy of those methods relies on a sufficient amount of high-quality training data, which itself can be expensive to obtain due to the requirement of domain knowledge and subject matter expertise. This paper is to resolve this dilemma by answering two questions: (1) provided with a limited number of reliable labels, can we use them to generate more synthetic labels; (2) Can we use those synthetic labels to improve the detectability? Among all the existing generative models, the generative adversarial network (GAN) shows its supreme capability in generating high-quality synthetic samples in multiple domains. We designed our model based on GAN. In particular, we studied several different network structures. By comparing the generated results, our GAN-based generative model yields the highest quality. We further combine the dataset with synthetic samples generated by our generative model and show that the detectability of our earthquake classification model is significantly improved than the one trained without augmenting the training set.

-- This paper proposes a framework which is able to generate a sequence of three-dimensional human dance poses for a given music. The proposed framework consists of three components: a music feature encoder, a pose generator, and a music genre classifier . We focus on integrating these components for generating a realistic 3D human dancing move from music, which can be applied to artificial agents and humanoid robots. The trained dance pose generator, which is a generative autoregressive model, is able to synthesize a dance sequence longer than 5,000 pose frames. Experimental results of generated dance sequences from various songs show how the proposed method generates humanlike dancing move to a given music. In addition, a generated 3D dance sequence is applied to a humanoid robot, showing that the proposed framework can make a robot to dance just by listening to music. Dance is one of the most important form of performing arts that having been emerged in all known cultures. As one of the specific subcategory of under theatrical dance, choreography associated with music is also one of the most popular forms that have usually been designed and physically performed by professional choreographers.

Unsupervised learning requiring only raw data is not only a fundamental function of the cerebral cortex, but also a foundation for a next generation of artificial neural networks. However, a unified theoretical framework to treat sensory inputs, synapses and neural activity together is still lacking. The computational obstacle originates from the discrete nature of synapses, and complex interactions among these three essential elements of learning. Here, we propose a variational mean-field theory in which only the distribution of synaptic weight is considered. The unsupervised learning can then be decomposed into two interwoven steps: a maximization step is carried out as a gradient ascent of the lower-bound on the data log-likelihood, and an expectation step is carried out as a message passing procedure on an equivalent or dual neural network whose parameter is specified by the variational parameter of the weight distribution. Therefore, our framework explains how data (or sensory inputs), synapses and neural activities interact with each other to achieve the goal of extracting statistical regularities in sensory inputs. This variational framework is verified in restricted Boltzmann machines with planted synaptic weights and learning handwritten digits.

Computer-assisted synthesis planning aims to help chemists find better reaction pathways faster. Finding viable and short pathways from sugar molecules to value-added chemicals can be modeled as a retrosynthesis planning problem with a catalyst allowed. This is a crucial step in efficient biomass conversion. The traditional computational chemistry approach to identifying possible reaction pathways involves computing the reaction energies of hundreds of intermediates, which is a critical bottleneck in silico reaction discovery. Deep reinforcement learning has shown in other domains that a well-trained agent with little or no prior human knowledge can surpass human performance. While some effort has been made to adapt machine learning techniques to the retrosynthesis planning problem, value-added chemical discovery presents unique challenges. Specifically, the reaction can occur in several different sites in a molecule, a subtle case that has never been treated in previous works. With a more versatile formulation of the problem as a Markov decision process, we address the problem using deep reinforcement learning techniques and present promising preliminary results.

Recently very deep transformers start showing outperformed performance to traditional bi-directional long short-term memory networks by a large margin. However, to put it into production usage, inference computation cost and latency are still serious concerns in real scenarios. In this paper, we study a novel non-autoregressive transformers structure for speech recognition, which is originally introduced in machine translation. During training input tokens fed to the decoder are randomly replaced by a special mask token. The network is required to predict those mask tokens by taking both context and input speech into consideration. During inference, we start from all mask tokens and the network gradually predicts all tokens based on partial results. We show this framework can support different decoding strategies, including traditional left-to-right. A new decoding strategy is proposed as an example, which starts from the easiest predictions to difficult ones. Some preliminary results on Aishell and CSJ benchmarks show the possibility to train such a non-autoregressive network for ASR. Especially in Aishell, the proposed method outperformed Kaldi nnet3 and chain model setup and is quite closed to the performance of the start-of-the-art end-to-end model.

Ensemble learning is a standard approach to building machine learning systems that capture complex phenomena in real-world data. An important aspect of these systems is the complete and valid quantification of model uncertainty. We introduce a Bayesian nonparametric ensemble (BNE) approach that augments an existing ensemble model to account for different sources of model uncertainty. BNE augments a model's prediction and distribution functions using Bayesian nonparametric machinery. It has a theoretical guarantee in that it robustly estimates the uncertainty patterns in the data distribution, and can decompose its overall predictive uncertainty into distinct components that are due to different sources of noise and error. We show that our method achieves accurate uncertainty estimates under complex observational noise, and illustrate its real-world utility in terms of uncertainty decomposition and model bias detection for an ensemble in predict air pollution exposures in Eastern Massachusetts, USA.

ABSTRACT In this work, we propose a new recurrent autoencoder architecture, termed Feedback Recurrent AutoEncoder (FRAE), for online compression of sequential data with temporal dependency. The recurrent structure of FRAE is designed to efficiently extract the redundancy along the time dimension and allows a compact discrete representation of the data to be learned. We demonstrate its effectiveness in speech spectrogram compression. Specifically, we show that the FRAE, paired with a powerful neural vocoder, can produce high-quality speech waveforms at a low, fixed bitrate. We further show that by adding a learned prior for the latent space and using an entropy coder, we can achieve an even lower variable bitrate.

Local differential privacy (LDP) is a model where users send privatized data to an untrusted central server whose goal it to solve some data analysis task. In the non-interactive version of this model the protocol consists of a single round in which a server sends requests to all users then receives their responses. This version is deployed in industry due to its practical advantages and has attracted significant research interest. Our main result is an exponential lower bound on the number of samples necessary to solve the standard task of learning a large-margin linear separator in the non-interactive LDP model. Via a standard reduction this lower bound implies an exponential lower bound for stochastic convex optimization and specifically, for learning linear models with a convex, Lipschitz and smooth loss. These results answer the questions posed in \citep{SmithTU17,DanielyF18}. Our lower bound relies on a new technique for constructing pairs of distributions with nearly matching moments but whose supports can be nearly separated by a large margin hyperplane. These lower bounds also hold in the model where communication from each user is limited and follow from a lower bound on learning using non-adaptive \emph{statistical queries}.