**NoisyNet-A3C** is a modification of [A3C](https://paperswithcode.com/method/a3c) that utilises noisy linear layers for exploration instead of 
$\epsilon$-greedy exploration as in the original [DQN](https://paperswithcode.com/method/dqn) formulation.

NoisyNet-A3C

Noisy Networks for Exploration

**Soft Actor Critic (Autotuned Temperature** is a modification of the [SAC](https://paperswithcode.com/method/soft-actor-critic) reinforcement learning algorithm. [SAC](https://paperswithcode.com/method/sac) can suffer from brittleness to the temperature hyperparameter. Unlike in conventional reinforcement learning, where the optimal policy is independent of scaling of the reward function, in maximum entropy reinforcement learning the scaling factor has to be compensated by the choice a of suitable temperature, and a sub-optimal temperature can drastically degrade performance. To resolve this issue, SAC with Autotuned Temperature has an automatic gradient-based temperature tuning method that adjusts the expected entropy over the visited states to match a target value.

Soft Actor-Critic (Autotuned Temperature)

Soft Actor-Critic Algorithms and Applications

Please enter a description about the method here

Enabling Retrain-free Deep Neural Network Pruning using Surrogate Lagrangian Relaxation

**Hierarchical Style Disentanglement**, or **HiSD**,  aims to disentangle different styles in image-to-image translation models. It organizes the labels into a hierarchical structure, where independent tags, exclusive attributes, and disentangled styles are allocated from top to bottom. To make the styles identified to the tags and attributes, the authors carefully redesign the modules, phases, and objectives.

HiSD

Image-to-image Translation via Hierarchical Style Disentanglement

Image Scale Augmentation is an augmentation technique where we randomly pick the short size of a image within a dimension range. One use case of this augmentation technique is in object detectiont asks.

Image Scale Augmentation

**Multi-View Entity Representations**, or **MuVER**, is an approach for entity retrieval that constructs multi-view representations for entity descriptions and approximates the optimal view for mentions via a heuristic searching method. It matches a mention to the appropriate entity by comparing it with entity descriptions. Motivated by the fact that mentions with different contexts correspond to different parts in descriptions, multi-view representations are constructed for each description. Specifically, we segment a description into several sentences. We refer to each sentence as a view $v$, which contains partial information, to form a view set $\mathcal{V}$ of the entity $e$. The Figure illustrates an example that constructs a view set $\mathcal{V}$ for “Kobe Bryant”.

MuVER

MuVER: Improving First-Stage Entity Retrieval with Multi-View Entity Representations

**Bridge-net** is an audio model block used in the [ClariNet](https://paperswithcode.com/method/clarinet) text-to-speech architecture. Bridge-net maps frame-level hidden representation to sample-level through several [convolution](https://paperswithcode.com/method/convolution) blocks and [transposed convolution](https://paperswithcode.com/method/transposed-convolution) layers interleaved with softsign non-linearities.

Bridge-net

ClariNet: Parallel Wave Generation in End-to-End Text-to-Speech

**Expected Sarsa** is like [Q-learning](https://paperswithcode.com/method/q-learning) but instead of taking the maximum over next state-action pairs, we use the expected value, taking into account how likely each action is under the current policy.

$$Q\left(S\_{t}, A\_{t}\right) \leftarrow Q\left(S\_{t}, A\_{t}\right) + \alpha\left[R_{t+1} + \gamma\sum\_{a}\pi\left(a\mid{S\_{t+1}}\right)Q\left(S\_{t+1}, a\right) - Q\left(S\_{t}, A\_{t}\right)\right] $$

Except for this change to the update rule, the algorithm otherwise follows the scheme of Q-learning. It is more computationally expensive than [Sarsa](https://paperswithcode.com/method/sarsa) but it eliminates the variance due to the random selection of $A\_{t+1}$.

Source: Sutton and Barto, Reinforcement Learning, 2nd Edition

Expected Sarsa

**AdamW** is a stochastic optimization method that modifies the typical implementation of weight decay in [Adam](https://paperswithcode.com/method/adam), by decoupling [weight decay](https://paperswithcode.com/method/weight-decay) from the gradient update. To see this, $L\_{2}$ regularization in Adam is usually implemented with the below modification where $w\_{t}$ is the rate of the weight decay at time $t$:

$$ g\_{t} = \nabla{f\left(\theta\_{t}\right)} + w\_{t}\theta\_{t}$$

while AdamW adjusts the weight decay term to appear in the gradient update:

$$ \theta\_{t+1, i} = \theta\_{t, i} - \eta\left(\frac{1}{\sqrt{\hat{v}\_{t} + \epsilon}}\cdot{\hat{m}\_{t}} + w\_{t, i}\theta\_{t, i}\right), \forall{t}$$

AdamW

Decoupled Weight Decay Regularization

This is **Graph Transformer** method, proposed as a generalization of [Transformer](https://paperswithcode.com/method/transformer) Neural Network architectures, for arbitrary graphs.

Compared to the original Transformer, the highlights of the presented architecture are:

- The attention mechanism is a function of neighborhood connectivity for each node in the graph.  
- The position encoding is represented by Laplacian eigenvectors, which naturally generalize the sinusoidal positional encodings often used in NLP.  
- The [layer normalization](https://paperswithcode.com/method/layer-normalization) is replaced by a [batch normalization](https://paperswithcode.com/method/batch-normalization) layer.  
- The architecture is extended to have edge representation, which can be critical to tasks with rich information on the edges, or pairwise interactions (such as bond types in molecules, or relationship type in KGs. etc).

Graph Transformer

A Generalization of Transformer Networks to Graphs

**YOLOv2**, or [**YOLO9000**](https://www.youtube.com/watch?v=QsDDXSmGJZA), is a single-stage real-time object detection model. It improves upon [YOLOv1](https://paperswithcode.com/method/yolov1) in several ways, including the use of [Darknet-19](https://paperswithcode.com/method/darknet-19) as a backbone, [batch normalization](https://paperswithcode.com/method/batch-normalization), use of a high-resolution classifier, and the use of anchor boxes to predict bounding boxes, and more.

YOLOv2

YOLO9000: Better, Faster, Stronger

**PVT**, or **Pyramid Vision Transformer**, is a type of [vision transformer](https://paperswithcode.com/methods/category/vision-transformer) that utilizes a pyramid structure to make it an effective backbone for dense prediction tasks. Specifically it allows for more fine-grained inputs (4 x 4 pixels per patch) to be used, while simultaneously shrinking the sequence length of the Transformer as it deepens - reducing the computational cost. Additionally, a [spatial-reduction attention](https://paperswithcode.com/method/spatial-reduction-attention) (SRA) layer is used to further reduce the resource consumption when learning high-resolution features.

The entire model is divided into four stages, each of which is comprised of a patch embedding layer and a $\mathcal{L}\_{i}$-layer Transformer encoder. Following a pyramid structure, the output resolution of the four stages progressively shrinks from high (4-stride) to low (32-stride).

Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction without Convolutions

**CheXNet** is a 121-layer [DenseNet](https://paperswithcode.com/method/densenet) trained on ChestX-ray14 for pneumonia detection.

CheXNet

CheXNet: Radiologist-Level Pneumonia Detection on Chest X-Rays with Deep Learning

Estimation statistics is a data analysis framework that uses a combination of effect sizes, confidence intervals, precision planning, and meta-analysis to plan experiments, analyze data and interpret results. It is distinct from null hypothesis significance testing (NHST), which is considered to be less informative. The primary aim of estimation methods is to report an effect size (a point estimate) along with its confidence interval, the latter of which is related to the precision of the estimate. The confidence interval summarizes a range of likely values of the underlying population effect. Proponents of estimation see reporting a P value as an unhelpful distraction from the important business of reporting an effect size with its confidence intervals, and believe that estimation should replace significance testing for data analysis.

Estimation Statistics

**Sharpness-Aware Minimization**, or **SAM**, is a procedure that improves model generalization by simultaneously minimizing loss value and loss sharpness. SAM functions by seeking parameters that lie in neighborhoods having uniformly low loss value (rather than parameters that only themselves have low loss value).

Sharpness-Aware Minimization

Sharpness-Aware Minimization for Efficiently Improving Generalization

The Fast-YOLOv4-SmallObj model is a modified version of Fast-[YOLOv4](https://paperswithcode.com/method/yolov4) to improve the detection of small objects. Seven layers were added so that it predicts bounding boxes at 3 different scales instead of 2.

Fast-YOLOv4-SmallObj

Towards Image-based Automatic Meter Reading in Unconstrained Scenarios: A Robust and Efficient Approach

A **SqueezeNeXt Block** is a two-stage bottleneck module used in the [SqueezeNeXt](https://paperswithcode.com/method/squeezenext) architecture to reduce the number of input channels to the 3 × 3 [convolution](https://paperswithcode.com/method/convolution). We decompose with separable convolutions to further reduce the number of parameters (orange parts), followed by a 1 × 1 expansion module.

SqueezeNeXt Block

SqueezeNext: Hardware-Aware Neural Network Design

**Position-Sensitive RoIAlign** is a positive sensitive version of [RoIAlign](https://paperswithcode.com/method/roi-align) - i.e. it performs selective alignment, allowing for the learning of position-sensitive region of interest aligning.

Viet-Anh on Software

AI Glossary

Tech Terminology