Revisiting $(\epsilon, \gamma, \tau)$-similarity learning for domain adaptation

Neural Information Processing Systems

Similarity learning is an active research area in machine learning that tackles the problem of finding a similarity function tailored to an observable data sample in order to achieve efficient classification. This learning scenario has been generally formalized by the means of a (ɛ, γ, τ) good similarity learning framework in the context of supervised classification and has been shown to have strong theoretical guarantees. In this paper, we propose to extend the theoretical analysis of similarity learning to the domain adaptation setting, a particular situation occurring when the similarity is learned and then deployed on samples following different probability distributions. We give a new definition of an (ɛ, γ) good similarity for domain adaptation and prove several results quantifying the performance of a similarity function on a target domain after it has been trained on a source domain. We particularly show that if the source distribution dominates the target one, then principally new domain adaptation learning bounds can be proved.



Improving Neural Program Synthesis with Inferred Execution Traces

Neural Information Processing Systems

The task of program synthesis, or automatically generating programs that are consistent with a provided specification, remains a challenging task in artificial intelligence. As in other fields of AI, deep learning-based end-to-end approaches have made great advances in program synthesis. However, compared to other fields such as computer vision, program synthesis provides greater opportunities to explicitly exploit structured information such as execution traces. While execution traces can provide highly detailed guidance for a program synthesis method, they are more difficult to obtain than more basic forms of specification such as input/output pairs. Therefore, we use the insight that we can split the process into two parts: infer traces from input/output examples, then infer programs from traces. Our application of this idea leads to state-of-the-art results in program synthesis in the Karel domain, improving accuracy to 81.3% from the 77.12% of prior work.








Neural Interaction Transparency (NIT): Disentangling Learned Interactions for Improved Interpretability

Neural Information Processing Systems

Neural networks are known to model statistical interactions, but they entangle the interactions at intermediate hidden layers for shared representation learning. We propose a framework, Neural Interaction Transparency (NIT), that disentangles the shared learning across different interactions to obtain their intrinsic lower-order and interpretable structure. This is done through a novel regularizer that directly penalizes interaction order. We show that disentangling interactions reduces a feedforward neural network to a generalized additive model with interactions, which can lead to transparent models that perform comparably to the state-of-theart models. NIT is also flexible and efficient; it can learn generalized additive models with maximumK-order interactions by training onlyO(1) models.