Search Results - "Koppula, Skanda"

EcoFlow: Efficient Convolutional Dataflows on Low-Power Neural Network Accelerators by Orosa, Lois, Koppula, Skanda, Umuroglu, Yaman, Kanellopoulos, Konstantinos, Gomez-Luna, Juan, Blott, Michaela, Vissers, Kees, Mutlu, Onur

“…Dilated and transposed convolutions are widely used in modern convolutional neural networks (CNNs). These kernels are used extensively during CNN training and…”
Get full text

Efficient Visual Pretraining with Contrastive Detection by Henaff, Olivier J., Koppula, Skanda, Alayrac, Jean-Baptiste, van den Oord, Aaron, Vinyals, Oriol, Carreira, Joao

“…Self-supervised pretraining has been shown to yield powerful representations for transfer learning. These performance gains come at a large computational cost…”
Get full text

Accurate, Low-Latency Visual Perception for Autonomous Racing: Challenges, Mechanisms, and Practical Solutions by Strobel, Kieran, Zhu, Sibo, Chang, Raphael, Koppula, Skanda

“…Autonomous racing provides the opportunity to test safety-critical perception pipelines at their limit. This paper describes the practical challenges and…”
Get full text

Energy-Efficient Speaker Identification with Low-Precision Networks by Koppula, Skanda, Glass, James, Chandrakasan, Anantha P.

“…Power-consumption in small devices is dominated by off-chip memory accesses, necessitating small models that can fit in on-chip memory. In the task of…”
Get full text

Understanding Recurrent Neural State Using Memory Signatures by Koppula, Skanda, Sim, Khe Chai, Chin, Kean

“…We demonstrate a network visualization technique to analyze the recurrent state inside the LSTMs/GRUs used commonly in language and acoustic models…”
Get full text

Learning a CNN-based End-to-End Controller for a Formula SAE Racecar by Koppula, Skanda

“…We present a set of CNN-based end-to-end models for controls of a Formula SAE racecar, along with various benchmarking and visualization tools to understand…”
Get full text

Power-Based Side-Channel Attack for AES Key Extraction on the ATMega328 Microcontroller by Banerjee, Utsav, Ho, Lisa, Koppula, Skanda

“…We demonstrate the extraction of an AES secret key from flash memory on the ATMega328 microcontroller (the microcontroller used on the popular Arduino…”
Get full text

Robust prediction-based analysis for genome-wide association and expression studies by K Koppula, Skanda, Zollanvari, Amin, An, Ning, Alterovitz, Gil

“…Here we describe a prediction-based framework to analyze omic data and generate models for both disease diagnosis and identification of cellular pathways which…”
Get full text

TAPVid-3D: A Benchmark for Tracking Any Point in 3D by Koppula, Skanda, Rocco, Ignacio, Yang, Yi, Heyward, Joe, Carreira, João, Zisserman, Andrew, Brostow, Gabriel, Doersch, Carl

“…We introduce a new benchmark, TAPVid-3D, for evaluating the task of long-range Tracking Any Point in 3D (TAP-3D). While point tracking in two dimensions (TAP)…”
Get full text

Applying the Residue Number System to Network Inference by Abdelhamid, Mohamed, Koppula, Skanda

“…This work explores the lesser studied objective of optimizing the multiply-and-accumulates executed during evaluation of the network. In particular, we propose…”
Get full text

Memory Consolidation Enables Long-Context Video Understanding by Balažević, Ivana, Shi, Yuge, Papalampidi, Pinelopi, Chaabouni, Rahma, Koppula, Skanda, Hénaff, Olivier J

“…Most transformer-based video encoders are limited to short temporal contexts due to their quadratic complexity. While various attempts have been made to extend…”
Get full text

A Simple Recipe for Contrastively Pre-Training Video-First Encoders Beyond 16 Frames by Papalampidi, Pinelopi, Koppula, Skanda, Pathak, Shreya, Chiu, Justin, Heyward, Joe, Patraucean, Viorica, Shen, Jiajun, Miech, Antoine, Zisserman, Andrew, Nematzdeh, Aida

“…Understanding long, real-world videos requires modeling of long-range visual dependencies. To this end, we explore video-first architectures, building on the…”
Get full text

Lossless Adaptation of Pretrained Vision Models For Robotic Manipulation by Sharma, Mohit, Fantacci, Claudio, Zhou, Yuxiang, Koppula, Skanda, Heess, Nicolas, Scholz, Jon, Aytar, Yusuf

“…Recent works have shown that large models pretrained on common visual learning tasks can provide useful representations for a wide range of specialized…”
Get full text

A Simple Recipe for Contrastively Pre-training Video-First Encoders Beyond 16 Frames by Papalampidi, Pinelopi, Koppula, Skanda, Pathak, Shreya, Chiu, Justin, Heyward, Joe, Patraucean, Viorica, Shen, Jiajun, Miech, Antoine, Zisserman, Andrew, Nematzdeh, Aida

“…Understanding long, real-world videos requires modeling of long-range visual dependencies. To this end, we explore video-first architectures, building on the…”
Get full text

Where Should I Spend My FLOPS? Efficiency Evaluations of Visual Pre-training Methods by Koppula, Skanda, Li, Yazhe, Shelhamer, Evan, Jaegle, Andrew, Parthasarathy, Nikhil, Arandjelovic, Relja, Carreira, João, Hénaff, Olivier

“…Self-supervised methods have achieved remarkable success in transfer learning, often achieving the same or better accuracy than supervised pre-training. Most…”
Get full text

BootsTAP: Bootstrapped Training for Tracking-Any-Point by Doersch, Carl, Luc, Pauline, Yang, Yi, Gokay, Dilara, Koppula, Skanda, Gupta, Ankush, Heyward, Joseph, Rocco, Ignacio, Goroshin, Ross, Carreira, João, Zisserman, Andrew

“…To endow models with greater understanding of physics and motion, it is useful to enable them to perceive how solid surfaces move and deform in real scenes…”
Get full text

Accurate, Low-Latency Visual Perception for Autonomous Racing:Challenges, Mechanisms, and Practical Solutions by Strobel, Kieran, Zhu, Sibo, Chang, Raphael, Koppula, Skanda

“…Autonomous racing provides the opportunity to test safety-critical perception pipelines at their limit. This paper describes the practical challenges and…”
Get full text

Understanding Recurrent Neural State Using Memory Signatures by Koppula, Skanda, Sim, Khe Chai, Chin, Kean

“…We demonstrate a network visualization technique to analyze the recurrent state inside the LSTMs/GRUs used commonly in language and acoustic models…”
Get full text

Object discovery and representation networks by Hénaff, Olivier J, Koppula, Skanda, Shelhamer, Evan, Zoran, Daniel, Jaegle, Andrew, Zisserman, Andrew, Carreira, João, Arandjelović, Relja

“…The promise of self-supervised learning (SSL) is to leverage large amounts of unlabeled data to solve complex tasks. While there has been excellent progress…”
Get full text

EcoFlow: Efficient Convolutional Dataflows for Low-Power Neural Network Accelerators by Orosa, Lois, Koppula, Skanda, Umuroglu, Yaman, Kanellopoulos, Konstantinos, Gomez-Luna, Juan, Blott, Michaela, Vissers, Kees, Mutlu, Onur

“…Dilated and transposed convolutions are widely used in modern convolutional neural networks (CNNs). These kernels are used extensively during CNN training and…”
Get full text