Alessandro Achille PhD student
University of California, Los Angeles
achille at cs.ucla.edu
alexachi
Since Fall 2015, I am a PhD student at the Computer Science department of UCLA, working with Prof. Stefano Soatto in the Vision Lab. I have also been a research scientist intern at Deep Mind since November 2017. My research interests include information theory, variational inference, representation learning, deep learning and their applications to computer vision.
Before coming to UCLA, I obtained a Master in Pure Math at the Scuola Normale Superiore and the University of Pisa, where I studied model theory, algebraic topology, and their intersection with Prof. Alessandro Berarducci. During that period, I have also been a visiting student at the University of Leeds Math department.
Publications

Critical Learning Periods in Deep Neural NetworksArXiv preprint
@ARTICLE{achille2017critical, author = {{Achille}, A. and {Rovere}, M. and {Soatto}, S.}, title = "{Critical Learning Periods in Deep Neural Networks}", journal = {ArXiv eprints}, archivePrefix = "arXiv", eprint = {1711.08856}, primaryClass = "cs.LG", keywords = {Computer Science  Learning, Quantitative Biology  Neurons and Cognition, Statistics  Machine Learning}, year = 2017, month = nov, }

A Separation Principle for Control in the Age of Deep LearningAnnual Reviews of Control, Robotics and Autonomous Systems, 2018
@article{achille2017separation, author = { Alessandro Achille and Stefano Soatto}, title = {A Separation Principle for Control in the Age of Deep Learning}, journal = {Annual Review of Control, Robotics, and Autonomous Systems}, volume = {1}, number = {1}, pages = {null}, year = {2018}, doi = {10.1146/annurevcontrol060117105140}, URL = { https://doi.org/10.1146/annurevcontrol060117105140 }, eprint = { https://doi.org/10.1146/annurevcontrol060117105140 } }

Emergence of Invariance and Disentangling in Deep RepresentationsProceedings of the ICML 2017 Workshop on Principled Approaches to Deep Learning
@ARTICLE{achille2017emergence, author = {{Achille}, A. and {Soatto}, S.}, title = "{Emergence of Invariance and Disentangling in Deep Representations}", journal = {Proceedings of the ICML Workshop on Principled Approaches to Deep Learning}, eprint = {1706.01350}, primaryClass = "cs.LG", keywords = {Computer Science  Learning, Computer Science  Artificial Intelligence, Statistics  Machine Learning}, year = 2017 }

Information Dropout: learning optimal representations through noisy computationTransactions on Pattern Analysis and Machine Intelligence (PAMI)
@ARTICLE{achille2018information, author={A. Achille and S. Soatto}, journal={IEEE Transactions on Pattern Analysis and Machine Intelligence}, title={Information Dropout: Learning Optimal Representations Through Noisy Computation}, year={2018}, volume={PP}, number={99}, pages={11}, keywords={Bayes methods;Information theory;Machine learning;Neural networks;Noise measurement;Training;Representation learning;deep learning;information bottleneck;invariants;minimality;nuisances}, doi={10.1109/TPAMI.2017.2784440}, ISSN={01628828}, month={},} }

A VietorisSmale mapping theorem for the homotopy of hyperdefinable setsSelecta Mathematica
@article{achille2018a, author = {Achille, Alessandro and Berarducci, Alessandro}, year = {2018}, title = {A VietorisSmale mapping theorem for the homotopy of hyperdefinable sets}, journal = {Selecta Mathematica}, issn = {10221824}, doi = {10.1007/s0002901804133}, month = {4}, pages = {129}, url = {http:https://doi.org/10.1007/s0002901804133}, abstract = {Results of Smale (1957) and Dugundji (1969) allow to compare the homotopy groups of two topological spaces X and Y whenever a map f:XâY with strong connectivity conditions on the fibers is given. We can apply similar techniques to compare the homotopy of spaces living in different categories, for instance an abelian variety over an algebraically closed field, and a real torus. More generally, working in ominimal expansions of fields, we compare the ominimal homotopy of a definable set X with the homotopy of some of its bounded hyperdefinable quotients X/E. Under suitable assumption, we show that pi_n^def(X)=pi_n(X/E) and dim(X)=dim_R(X/E). As a special case, given a definably compact group, we obtain a new proof of Pillay's group conjecture dim(G)=dim_R(G/G00) largely independent of the group structure of G. We also obtain different proofs of various comparison results between classical and ominimal homotopy.} }