Dan Feldman

Dr. Dan Feldman

Robotics & Big Data Lab, Computer Science Department, University of Haifa, Haifa 3498838, Israel
orcid

Research Focus & Metrics

Academic Career Length

14Years (2013-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 October 2019 in WIREs Data Mining and Knowledge Discovery
Verified
Journal Article
Closed access
PDF file
DFDan Feldman
Dan Feldman

30 August 2021 in Moral Philosophy and Politics
Verified
Journal Article
Open access
NENir Eisikovits
Nir Eisikovits
DFDan Feldman
Dan Feldman
We argue that the growing prevalence of statistical machine learning in everyday decision making – from creditworthiness to police force allocation – effectively replaces many of our humdrum practical judgments and that this will eventually undermine our capacityfor making such judgments. We lean on Aristotle’s famous account of howphronesisand moral virtues develop to make our case. If Aristotle is right that the habitual exercise of practical judgment allows us to incrementally hone virtues, and if AI saves us time by taking over some of those practical judgments, or if its pattern recognition capacities are very good at learning that kind of behavior – we risk innovating ourselves out of moral competence with the introduction of AI.

15 October 2018
Verified
Conference Paper
Open access
PDF file
AAAdi Akavia
Adi Akavia
DFDan Feldman
Dan Feldman
HSHayim Shaul
Hayim Shaul

24 June 2021 in IEEE Transactions on Neural Networks and Learning Systems
Verified
Journal Article
Closed access
BMBen Mussay
Ben Mussay
DFDan Feldman
Dan Feldman
SZSamson Zhou
Samson Zhou
VBVladimir Braverman
Vladimir Braverman
MOMargarita Osadchy
Margarita Osadchy
Model compression is crucial for the deployment of neural networks on devices with limited computational and memory resources. Many different methods show comparable accuracy of the compressed model and similar compression rates. However, the majority of the compression methods are based on heuristics and offer no worst case guarantees on the tradeoff between the compression rate and the approximation error for an arbitrarily new sample. We propose the first efficient structured pruning algorithm with a provable tradeoff between its compression rate and the approximation error for any future test sample. Our method is based on the coreset framework, and it approximates the output of a layer of neurons/filters by a coreset of neurons/filters in the previous layer and discards the rest. We apply this framework in a layer-by-layer fashion from the bottom to the top. Unlike previous works, our coreset is data-independent, meaning that it provably guarantees the accuracy of the function for any input x ∈ Rd, including an adversarial one.

20 March 2013 in Journal of Mathematical Imaging and Vision
Verified
Journal Article
Closed access
PDF file
DFDan Feldman
Dan Feldman
MFMicha Feigin
Micha Feigin
NSNir Sochen
Nir Sochen

30 June 2016
Verified
Conference Paper
Open access
ABArtem Barger
Artem Barger
DFDan Feldman
Dan Feldman
We provide the first streaming algorithm for computing a provable approximation to the k-means of sparse Big Data. Here, sparse Big Data is a stream of n vectors in ℝd, where each vector has O(1) non-zeroes entries and possibly d ≥ n. E.g., adjacency matrix of a graph, web-links, social network, document-terms, or image-features matrices. Our streaming algorithm stores at most log n · kO(1) input points in memory. If the stream is distributed among M machines, the running time reduces by a factor of M, while communicating a total of M · kO(1) (sparse) input points between the machines. Our main contribution is a deterministic algorithm for computing a sparse (k, ∊)-coreset, which is a weighted subset of kO(1) input points that approximates the sum of squared distances from the n input points to every set of k centers, up to (1 ± ∊) factor, for any given constant ε > 0. This is the first such coreset of size independent of both d and n. Our experimental results show how our algorithm can bs used to boost the performance of any given k-means heuristics, even in the off-line setting. Open access to our implementation is also provided.