This paper introduces transductive consensus network (TCNs), as an extension of a consensus network (CN), for semi-supervised learning. TCN does multi-modal classification based on a few available labels by urging the {\em interpretations} of different modalities to resemble each other. We formulate the multi-modal, semi-supervised learning problem, put forward TCN for multi-modal semi-supervised learning task, and its several variants. To understand the mechanisms of TCN, we formulate the {\em similarity} of the interpretations as the negative relative Jensen-Shannon divergence, and show that a consensus state beneficial for classification desires a stable but not perfect similarity between the interpretations. We show the performances of TCN are better than best benchmark algorithms given only 20 and 80 labeled samples on Bank Marketing and the DementiaBank dataset respectively, and align with their performances given more labeled samples.

The standard margin-based structured prediction commonly uses a maximum loss over all possible structured outputs [23, 1, 5, 22]. The large-margin formulation including latent variables [27, 18] not only results in a non-convex formulation but also increases the search space by a factor of the size of the latent space. Recent work [11] has proposed the use of the maximum loss over random structured outputs sampled independently from some proposal distribution, with theoretical guarantees. We extend this work by including latent variables. We study a new family of loss functions under Gaussian perturbations and analyze the effect of the latent space on the generalization bounds. We show that the non-convexity of learning with latent variables originates naturally, as it relates to a tight upper bound of the Gibbs decoder distortion with respect to the latent space. Finally, we provide a formulation using random samples that produces a tighter upper bound of the Gibbs decoder distortion up to a statistical accuracy, which enables a faster evaluation of the objective function. We illustrate the method with synthetic experiments and a computer vision application.

GANS are powerful generative models that are able to model the manifold of natural images. We leverage this property to perform manifold regularization by approximating the Laplacian norm using a Monte Carlo approximation that is easily computed with the GAN. When incorporated into the feature-matching GAN of Improved GAN, we achieve state-of-the-art results for GAN-based semi-supervised learning on the CIFAR-10 dataset, with a method that is significantly easier to implement than competing methods.

We consider the task of training classifiers without labels. We propose a weakly supervised method---adversarial label learning---that trains classifiers to perform well against an adversary that chooses labels for training data. The weak supervision constrains what labels the adversary can choose. The method therefore minimizes an upper bound of the classifier's error rate using projected primal-dual subgradient descent. Minimizing this bound protects against bias and dependencies in the weak supervision. Experiments on three real datasets show that our method can train without labels and outperforms other approaches for weakly supervised learning.

We give an algorithmically efficient version of the learner-to-compression scheme conversion in Moran and Yehudayoff (2016). In extending this technique to real-valued hypotheses, we also obtain an efficient regression-to-bounded sample compression converter. To our knowledge, this is the first general compressed regression result (regardless of efficiency or boundedness) guaranteeing uniform approximate reconstruction. Along the way, we develop a generic procedure for constructing weak real-valued learners out of abstract regressors; this may be of independent interest. In particular, this result sheds new light on an open question of H. Simon (1997). We show applications to two regression problems: learning Lipschitz and bounded-variation functions.

MAP perturbation models have emerged as a powerful framework for inference in structured prediction. Such models provide a way to efficiently sample from the Gibbs distribution and facilitate predictions that are robust to random noise. In this paper, we propose a provably polynomial time randomized algorithm for learning the parameters of perturbed MAP predictors. Our approach is based on minimizing a novel Rademacher-based generalization bound on the expected loss of a perturbed MAP predictor, which can be computed in polynomial time. We obtain conditions under which our randomized learning algorithm can guarantee generalization to unseen examples.

Computer vision is one of today's most exciting application fields of Machine Learning, From self-driving cars to Medical diagnosis, this has been widely used in various domains. This course will take you right from the essential concepts of statistical learning to help you with various algorithms to implement it with other OpenCV tasks. The course will also guide you through creating custom graphs and visualizations, and show you how to go from the raw data to beautiful visualizations. We will also build a machine learning system that can make a medical diagnosis. By the end of this course, you will be ready create your own ML system and will also be able to take on your own machine learning problems.

One of the first lessons you'll receive in machine learning is that there are two broad categories: supervised and unsupervised learning. Supervised learning is usually explained as the one to which you provide the correct answers, training data, and the machine learns the patterns to apply to new data. Unsupervised learning is (apparently) where the machine figures out the correct answer on its own. Supposedly, unsupervised learning can discover something new that has not been found in the data before. Supervised learning cannot do that.

Embedding methods which enforce a partial order or lattice structure over the concept space, such as Order Embeddings (OE) (Vendrov et al., 2016), are a natural way to model transitive relational data (e.g. entailment graphs). However, OE learns a deterministic knowledge base, limiting expressiveness of queries and the ability to use uncertainty for both prediction and learning (e.g. learning from expectations). Probabilistic extensions of OE (Lai and Hockenmaier, 2017) have provided the ability to somewhat calibrate these denotational probabilities while retaining the consistency and inductive bias of ordered models, but lack the ability to model the negative correlations found in real-world knowledge. In this work we show that a broad class of models that assign probability measures to OE can never capture negative correlation, which motivates our construction of a novel box lattice and accompanying probability measure to capture anticorrelation and even disjoint concepts, while still providing the benefits of probabilistic modeling, such as the ability to perform rich joint and conditional queries over arbitrary sets of concepts, and both learning from and predicting calibrated uncertainty. We show improvements over previous approaches in modeling the Flickr and WordNet entailment graphs, and investigate the power of the model.

Constand's testimony and that of five other accusers allowed prosecutors to uncloak Cosby -- who solidified his reputation as a family man by playing Dr. Cliff Huxtable on "The Cosby Show" -- as a manipulative predator who used his built-in trust to trick women into taking powerful intoxicants so he could violate them.