Standard few-shot experiments involve learning to efficiently match unseen samples by class. We claim that few-shot learning should be long term, assimilating knowledge for the future, without forgetting previous concepts. In the mammalian brain, the hippocampus is understood to play a significant role in this process, by learning rapidly and consolidating knowledge to the neocortex incrementally over a short period. In this research we tested whether an artificial hippocampal algorithm (AHA), could be used with a conventional Machine Learning (ML) model that learns incrementally analogous to the neocortex, to achieve one-shot learning both short and long term. The results demonstrated that with the addition of AHA, the system could learn in one-shot and consolidate the knowledge for the long term without catastrophic forgetting. This study is one of the first examples of using a CLS model of hippocampus to consolidate memories, and it constitutes a step toward few-shot continual learning.

We present a recurrent neural network memory that uses sparse coding to create a combinatoric encoding of sequential inputs. The network is trained using only local and immediate credit assignment. Despite this constraint, results are comparable to networks trained using deep backpropagation or BackProp Through Time (BPTT). With several examples, we show that the network can associate distant cause and effect in a discrete stochastic process, predict partially-observable higherorder sequences, and learn to generate many time-steps of video simulations. Typical memory consumption is 10-30x less than conventional RNNs, such as LSTM, trained by BPTT. One limitation of the memory is generalization to unseen input sequences. We additionally explore this limitation by measuring next-word prediction perplexity on the Penn Treebank dataset.

We present a probabilistic, salience-based mechanism for the interpretation of pointing gestures together with spoken utterances. Our formulation models dependencies between spatial and temporal aspects of gestures and features of objects. The results from our corpus-based evaluation show that the incorporation of pointing information improves interpretation accuracy. Index Terms: understanding spoken language and gesture, probabilistic approach 1.

Odour localisation in an enclosed area is difficult due to the formation of sectors of circulating airflow. Well-defined plumes do not exist, and reactive plume following may not be possible. Odour localisation has been partially achieved in this environment by using knowledge of airflow, and a search that relies on chemical sensing and reasoning. However the results are not specific, with the odour source only restricted to a broad area. This paper presents a solution to the problem by introducing a second search stage using visual sensing. It therefore comprises a bi-modal, two-stage search, with each stage exploiting complementary sensing modalities. This paper presents details of the method and experimental results