Christopher H. Bennett
@cbennett10
Followers
765
Following
12K
Media
266
Statuses
4K
Eclectic engineer & nanoscientist. researcher @ Sandia Labs. Previously: PhD #nanoelectronics @u_psud; software startup, @stanford ugrad. also❤️ :📸 🏃 🏔 🧬
Joined March 2008
My #PhD thesis on local #learning w emerging #ReRAM nanodevices is now available online & open-access :🔓🤖👨💻 #nanoelectronics #nanotechnology #neuromorphic #computing.
theses.hal.science
In the next era of distributed computing, brain-based computers that perform operations locally rather than in remote servers would be a major benefit in reducing global energy costs. A new generat...
1
5
23
Check out our pre-print on #analog fast fourier transforms (#FFTs) using large memory arrays: . Our work suggests a new application point for #neuromorphic systems. Notably, analog FFTs scale well, outperforming digital FFTs at large signal size.
arxiv.org
Edge devices are being deployed at increasing volumes to sense and act on information from the physical world. The discrete Fourier transform (DFT) is often necessary to make this sensed data...
0
1
5
RT @naturecomputes: In this new paper, led by @giovannimarchet, we present a unique theoretical result that provides guarantees for the con….
0
54
0
RT @JeffDean: Paper: TPUv4 system has an optically reconfigurable network to assemble groups of 4x4x4 chips like legos (4x4x12? 16x16x16?).….
0
116
0
RT @IOPneuromorphic: Part of this topical review from @UT_Dallas, @UTAustin, @usal & @SandiaLabs looks at magnetic….
0
5
0
RT @NatureElectron: Our January issue is now out, featuring substrate-less nanomesh receptors, non-reciprocal acoustoelectric microwave amp….
0
9
0
RT @NeelNanda5: Excited to announce that our work, Progress Measures for Grokking via Mechanistic Interpretability, has been accepted as a….
0
84
0
RT @patrickmineault: I gave a keynote last month at MAIN educational on the future of neuroAI, recording now available .
0
9
0
RT @boahen_k: My seCReT master plan to achieve #NeuralSupremacy is out! I would like to thank my editors at Nature….
0
16
0
RT @DamienQuerlioz: Discover our memristor-based Bayesian machine, just published in @NatureElectron ! This is our first completely integra….
0
41
0
We’ve made the cover in our new work exploring #domainwall #spintronic devices as a multi-functional #neuromorphic #computing platform 😁🧲Great collab with @NanoComputeLab !
advanced.onlinelibrary.wiley.com
Neuromorphic Computing This frontispiece imagines the use of the studied devices in a neural network crossbar array for brain-inspired computing. The devices function as artificial synapses at each...
1
2
8
RT @zen_fi: The mind is just the current thought, the body is the current sensation, the world is the current perception.
0
79
0
RT @zetalyrae: The fact we automated human creativity before manual drudgery is fascinating. It is almost enough to make you believe in a t….
0
64
0
RT @jbimaknee: The brain has true random number generators everywhere. Our computers don't. What if they did?.
advanced.onlinelibrary.wiley.com
Modern computing assumes that materials, devices, and circuits will be operated in a fully deterministic manner, providing predictable and reliable outputs, with any desired random behavior re-intr...
0
1
0
RT @Michael_J_Black: I asked #Galactica about some things I know about and I'm troubled. In all cases, it was wrong or biased but sounded r….
0
727
0
New article 📖: proposes a new approach for sampling magnetic tunnel junctions (#MTJs) thereby implementing a form of variational inference (an approximate flavor of #Bayesian #inference). This could enable out-of-distribution detection at the edge .
1
0
5
RT @ai_ngrosso: Convolutional structure learned from scratch in fully connected neural networks. A tale of non-gaussianity and cumulant ten….
pnas.org
Exploiting data invariances is crucial for efficient learning in both artificial and biological neural circuits. Understanding how neural networks ...
0
23
0
0
43
0
New #preprint up ! Great collab effort between SNL and @NanoComputeLab researchers on graphene #nanosynapses #nanoelectronics #neuromorphiccomputing #bioelectronics.
0
0
3