A generative recurrent neural network is quickly trained in an unsupervised manner to model popular reinforcement learning environments through compressed spatio-temporal representations. The world model's extracted features are fed into compact and simple policies trained by evolution, achieving state of the art results in various environments. We also train our agent entirely inside of an environment generated by its own internal world model, and transfer this policy back into the actual environment. Interactive version of this paper is available at https://worldmodels.github.io

A fundamental problem in program verification concerns inferring loop invariants. The problem is undecidable and even practical instances are challenging. Inspired by how human experts construct loop invariants, we propose a reasoning framework Code2Inv that constructs the solution by multi-step decision making and querying an external program graph memory block. By training with reinforcement learning, Code2Inv captures rich program features and avoids the need for ground truth solutions as supervision. Compared to previous learning tasks in domains with graph-structured data, it addresses unique challenges, such as a binary objective function and an extremely sparse reward that is given by an automated theorem prover only after the complete loop invariant is proposed. We evaluate Code2Inv on a suite of 133 benchmark problems and compare it to three state-of-the-art systems. It solves 106 problems compared to 73 by a stochastic search-based system, 77 by a heuristic search-based system, and 100 by a decision tree learning-based system. Moreover, the strategy learned can be generalized to new programs: compared to solving new instances from scratch, the pre-trained agent is more sample efficient in finding solutions.

A generative recurrent neural network is quickly trained in an unsupervised manner to model popular reinforcement learning environments through compressed spatio-temporal representations. The world model's extracted features are fed into compact and simple policies trained by evolution, achieving state of the art results in various environments. We also train our agent entirely inside of an environment generated by its own internal world model, and transfer this policy back into the actual environment. Interactive version of this paper is available at https://worldmodels.github.io

Beyond local convolution networks, we explore how to harness various external human knowledge for endowing the networks with the capability of semantic global reasoning. Rather than using separate graphical models (e.g. CRF) or constraints for modeling broader dependencies, we propose a new Symbolic Graph Reasoning (SGR) layer, which performs reasoning over a group of symbolic nodes whose outputs explicitly represent different properties of each semantic in a prior knowledge graph. To cooperate with local convolutions, each SGR is constituted by three modules: a) a primal local-to-semantic voting module where the features of all symbolic nodes are generated by voting from local representations; b) a graph reasoning module propagates information over knowledge graph to achieve global semantic coherency; c) a dual semantic-to-local mapping module learns new associations of the evolved symbolic nodes with local representations, and accordingly enhances local features. The SGR layer can be injected between any convolution layers and instantiated with distinct prior graphs. Extensive experiments show incorporating SGR significantly improves plain ConvNets on three semantic segmentation tasks and one image classification task. More analyses show the SGR layer learns shared symbolic representations for domains/datasets with the different label set given a universal knowledge graph, demonstrating its superior generalization capability.

Mixed inference such as the marginal MAP query (some variables marginalized by summation and others by maximization) is key to many prediction and decision models. It is known to be extremely hard; the problem is NPPP-complete while the decision problem for MAP is only NP-complete and the summation problem is #P-complete. Consequently, approximation anytime schemes are essential. In this paper, we show that the framework of heuristic AND/OR search, which exploits conditional independence in the graphical model, coupled with variational-based mini-bucket heuristics can be extended to this task and yield powerful state-of-the-art schemes. Specifically, we explore the complementary properties of best-first search for reducing the number of conditional sums and providing time-improving upper bounds, with depth-first search for rapidly generating and improving solutions and lower bounds. We show empirically that a class of solvers that interleaves depth-first with best-first schemes emerges as the most competitive anytime scheme.

Neural link predictors learn distributed representations of entities and relations in a knowledge graph. They are remarkably powerful in the link prediction and knowledge base completion tasks, mainly due to the learned representations that capture important statistical dependencies in the data. Recent works in the area have focused on either designing new scoring functions or incorporating extra information into the learning process to improve the representations. Yet the representations are mostly learned from the observed links between entities, ignoring commonsense or schema knowledge associated with the relations in the graph. A fundamental aspect of the topology of relational data is the cardinality information, which bounds the number of predictions given for a relation between a minimum and maximum frequency. In this paper, we propose a new regularisation approach to incorporate relation cardinality constraints to any existing neural link predictor without affecting their efficiency or scalability. Our regularisation term aims to impose boundaries on the number of predictions with high probability, thus, structuring the embeddings space to respect commonsense cardinality assumptions resulting in better representations. Experimental results on Freebase, WordNet and YAGO show that, given suitable prior knowledge, the proposed method positively impacts the predictive accuracy of downstream link prediction tasks.

Exploiting dependencies between labels is considered to be crucial for multi-label classification. Rules are able to expose label dependencies such as implications, subsumptions or exclusions in a human-comprehensible and interpretable manner. However, the induction of rules with multiple labels in the head is particularly challenging, as the number of label combinations which must be taken into account for each rule grows exponentially with the number of available labels. To overcome this limitation, algorithms for exhaustive rule mining typically use properties such as anti-monotonicity or decomposability in order to prune the search space. In the present paper, we examine whether commonly used multi-label evaluation metrics satisfy these properties and therefore are suited to prune the search space for multi-label heads.

It is a relatively mild scene in a documentary about the sexual predator who helped transform American politics. Back when he ran Fox News, Roger Ailes bought up his hometown paper, and in Divide and Conquer--now in theaters--the Putnam County News and Recorder's former copy editor describes what happened to her after she eventually quit the job. In the next few days, people she had messaged privately about Ailes on Facebook began finding out that he was looking into them. One even received a phone call: "This is Roger Ailes, and I hear you've been making threats about me." Ailes then quoted the friends' Facebook conversation, verbatim. In her interview with Divide and Conquer, the copy editor was clearly still shaken by the experience: "It was really terrifying--this feeling that there are really powerful people who live five minutes from me that are out to destroy me and my life."

We propose Deep Hierarchical Machine (DHM), a model inspired from the divide-and-conquer strategy while emphasizing representation learning ability and flexibility. A stochastic routing framework as used by recent deep neural decision/regression forests is incorporated, but we remove the need to evaluate unnecessary computation paths by utilizing a different topology and introducing a probabilistic pruning technique. We also show a specified version of DHM (DSHM) for efficiency, which inherits the sparse feature extraction process as in traditional decision tree with pixel-difference feature. To achieve sparse feature extraction, we propose to utilize sparse convolution operation in DSHM and show one possibility of introducing sparse convolution kernels by using local binary convolution layer. DHM can be applied to both classification and regression problems, and we validate it on standard image classification and face alignment tasks to show its advantages over past architectures.

In a letter addressed to her Democratic colleagues and members-elect over Thanksgiving weekend, House Democratic Leader Nancy Pelosi gushed over the "important reforms" the Democratic majority would make in its rules package for the 116th Congress. Among these captivating new changes, she wrote, the package would "establish a select committee to improve the operation of Congress, ensuring that we deliver in a manner that is transparent, bipartisan and unifying." Establishing a committee to further investigate a topic of concern, however, is a congressional leader's way of telling a particular bloc of members that have been seeking specific concrete action, "No, you will not be getting that." In this case, the group Pelosi was saying "no" to was the Democratic membership of the House Problem Solvers Caucus, a bipartisan, moderate caucus affiliated with the centrist political advocacy group No Labels. A group of nine of the caucus' Democratic members--down from 14 a few weeks ago--has been threatening to withhold its votes for Pelosi for speaker on the House floor in January unless she agrees to a series of a rules change that, in their words, would "allow for more transparency and bipartisan governing."