Learning to detect content-independent transformations from data is one of the central problems in biological and artificial intelligence. An example of such problem is unsupervised learning of a visual motion detector from pairs of consecutive video frames. Rao and Ruderman formulated this problem in terms of learning infinitesimal transformation operators (Lie group generators) via minimizing image reconstruction error. Unfortunately, it is difficult to map their model onto a biologically plausible neural network (NN) with local learning rules. Here we propose a biologically plausible model of motion detection. We also adopt the transformation-operator approach but, instead of reconstruction-error minimization, start with a similarity-preserving objective function.

This paper examines neural circuits for motion detection, which have been extensively modeled and are certainly of interest to the field. A nice connection is drawn with recent work on biologically plausible learning of similarity matching objective functions, which is also of great interest to the field. While the work draws a nice connection on timely topics, some of the conclusions and biological predictions were difficult for me to follow. While the paper is decently written, the presentation is dense and difficult to follow in several places. This is partly due to space constraints on the manuscript, but I nonetheless think there is room for improvement in sections 2.1, section 3, and section 4. Section 2.1 and the preceding paragraph would be easily understood by readers with a background in group theory (I imagine this would be mostly limited to physicists), but I don't think this level of sophistication is necessary to convey the main ideas of the paper.

This work considers similarity-preserving objective functions for learning to classify inputs with a temporal dimension. The authors propose a modification of the Lie algebra formulation of Ruderman and Rao, where the algorithm maximizes the similarity of transformation of inputs that are nearby in time rather than comparing inputs at the same time directly. While the scores given were worthy of acceptance, the enthusiasm of reviewers both in the body of the reviews and in the discussion was somewhat muted. My impression is that there were two main reasons for this. Thus, while I see no reason to contradict the recommendation of the reviewers that the paper be accepted, we expect the reviewers to address these points (and the clarity of the paper in general) in the camera ready version of the paper.

Learning to detect content-independent transformations from data is one of the central problems in biological and artificial intelligence. An example of such problem is unsupervised learning of a visual motion detector from pairs of consecutive video frames. Rao and Ruderman formulated this problem in terms of learning infinitesimal transformation operators (Lie group generators) via minimizing image reconstruction error. Unfortunately, it is difficult to map their model onto a biologically plausible neural network (NN) with local learning rules. Here we propose a biologically plausible model of motion detection.

Learning to detect content-independent transformations from data is one of the central problems in biological and artificial intelligence. An example of such problem is unsupervised learning of a visual motion detector from pairs of consecutive video frames. Rao and Ruderman formulated this problem in terms of learning infinitesimal transformation operators (Lie group generators) via minimizing image reconstruction error. Unfortunately, it is difficult to map their model onto a biologically plausible neural network (NN) with local learning rules. Here we propose a biologically plausible model of motion detection.