Learning from Observation (LfO) is highly useful for modeling behaviors through nonintrusive observation of some actor's performance. However, an actor's performance is often influenced by unobservable internal influences, such as emotions, agendas, and memory of past events. Therefore, new techniques are needed to infer the structure of these influences and their effect on an actor's decisions. In this paper, we propose a novel approach called Memory Composition Learning (MCL) for capturing one internal influence: memory of past events. We hypothesize that memory influences on a behavior can be modeled through parameterized memory features that can be learned from observation of traces of an actor's behavior; these memory features can then be presented as additional input to a performance modeling application. We demonstrate the efficacy of our approach in a simulated vacuum cleaner domain and show that hidden memory influences can be detected, modeled, and then used to improve machine learning performance.

The problem of attributing a deep network's prediction to its \emph{input/base} features is well-studied. We introduce the notion of \emph{conductance} to extend the notion of attribution to the understanding the importance of \emph{hidden} units. Informally, the conductance of a hidden unit of a deep network is the \emph{flow} of attribution via this hidden unit. We use conductance to understand the importance of a hidden unit to the prediction for a specific input, or over a set of inputs. We evaluate the effectiveness of conductance in multiple ways, including theoretical properties, ablation studies, and a feature selection task. The empirical evaluations are done using the Inception network over ImageNet data, and a sentiment analysis network over reviews. In both cases, we demonstrate the effectiveness of conductance in identifying interesting insights about the internal workings of these networks.

We study the problem of explaining a rich class of behavioral properties of deep neural networks. Distinctively, our influence-directed explanations approach this problem by peering inside the net- work to identify neurons with high influence on the property and distribution of interest using an axiomatically justified influence measure, and then providing an interpretation for the concepts these neurons represent. We evaluate our approach by training convolutional neural net- works on MNIST, ImageNet, Pubfig, and Diabetic Retinopathy datasets. Our evaluation demonstrates that influence-directed explanations (1) identify influential concepts that generalize across instances, (2) help extract the essence of what the network learned about a class, (3) isolate individual features the network uses to make decisions and distinguish related instances, and (4) assist in understanding misclassifications.

Diffusion of items occurs in social networks due to spreading of items through word of mouth and exogenous factors. These items may be news, products, videos, advertisements or contagious viruses. Previous research has studied diffusion process at both the macro and micro levels. The former models the number of item adopters in the diffusion process while the latter determines which individuals adopt item. In this paper, we establish a general probabilistic framework, which can be used to derive macro-level diffusion models, including the well known Bass Model (BM). Using this framework, we develop several other models considering the social network’s degree distribution coupled with the assumption of linear influence by neighboring adopters in the diffusion process. Through some evaluation on synthetic data, this paper shows that degree distribution actually changes during the diffusion process. We therefore introduce a multi-stage diffusion model to cope with variable degree distribution. By conducting experiments on both synthetic and real datasets, we show that our proposed diffusion models can recover the diffusion parameters from the observed diffusion data, which allows us to model diffusion with high accuracy.